Solar Photooxygenations for the Manufacturing of Fine Chemicals—Technologies and Applications

molecules

Review Solar Photooxygenations for the Manufacturing of Fine Chemicals—Technologies and Applications

Jayson S. Wau 1, Mark J. Robertson 1 and Michael Oelgemöller 1,2,*

1 College of Science and Engineering, James Cook University, Townsville, QLD 4811, Australia; [email protected] (J.S.W.); [email protected] (M.J.R.) 2 Department of Organic and Macromolecular Chemistry, Ghent University, Krijgslaan 281 S4, 9000 Gent, Belgium * Correspondence: [email protected]; Tel.: +61-7-4781-4543

1 Abstract: Photooxygenation reactions involving singlet oxygen ( O2) are utilized industrially as a mild and sustainable access to oxygenated products. Due to the usage of organic dyes as photosensitizers, these transformations can be successfully conducted using natural sunlight. Modern solar chemical reactors enable outdoor operations on the demonstration (multigram) to technical (multikilogram) scales and have subsequently been employed for the manufacturing of ﬁne chemicals such as fragrances or biologically active compounds. This review will highlight examples of solar photooxygenations for the manufacturing of industrially relevant target compounds and will discuss current challenges and opportunities of this sustainable methodology.

Keywords: solar chemistry; photooxygenations; solar reactor; chemical manufacturing

Citation: Wau, J.S.; Robertson, M.J.; Oelgemöller, M. Solar Photooxygenations for the 1. Introduction Manufacturing of Fine Photochemistry has recently seen a remarkable renaissance in synthetic organic chem- Chemicals—Technologies and istry [1–4]. This trend was sparked by the development of new photochemical transfor- Applications. Molecules 2021, 26, mations [5–8] as well as advances in photoreactor and light technologies [9–13]. However, 1685. https://doi.org/10.3390/ the large-scale realization of photochemical manufacturing processes is hindered by their molecules26061685 quantum yields and the significant installation and maintenance costs of artificial light sources [14,15]. As a result, industrial photochemistry is restricted to low-volume but Academic Editor: Axel G. Griesbeck high-value fine chemicals (such as fragrances, flavors or vitamins), or bulk but low-value chemicals (such as haloalkanes, oximes, sulfonyl chlorides and sulfonic acids) [14–19]. Received: 18 February 2021 Natural sunlight is regarded as an alternative and sustainable energy source for a range Accepted: 15 March 2021 of photochemical transformations [20–23]. Up until the early 20th century, photochemical Published: 17 March 2021 experiments were routinely performed by placing sealed flasks and tubes in direct sunlight [24–27]. Although this ‘flask in the sun’ approach is still followed today [28,29], the Publisher’s Note: MDPI stays neutral small scales and prolonged exposure times result in very low productivities of the desired with regard to jurisdictional claims in published maps and institutional affil- products. Recently, sunlight collectors and concentrators that were initially developed iations. for energy generation [30–33] have been modified for solar photochemical experimenta- tion [34]. Technical-scale solar degradation and detoxification processes have subsequently been successfully realized using these technologies [35–38]. In contrast, however, preparative solar chemistry for the manufacturing of chemicals is still comparably rare [39–47]. The main hurdles for a widespread implementation of solar synthesis are the discontinu- Copyright: © 2021 by the authors. ous availability and the low UV content (<400 nm) of just 3–5% of natural sunlight [48]. Licensee MDPI, Basel, Switzerland. Sunlight also consists largely of direct and diffuse radiation and their ratios vary with This article is an open access article distributed under the terms and the location and weather [49]. Solar manufacturing may thus be ideally performed in the conditions of the Creative Commons visible range (400–700 nm: 42–43% of solar spectrum [48]) for small annual production Attribution (CC BY) license (https:// targets (<100 tons/year) that do not require continuous, year-round operation under opti- 1 creativecommons.org/licenses/by/ mal solar light conditions. Photooxygenations with singlet oxygen ( O2) combine catalytic 4.0/).

Molecules 2021, 26, 1685. https://doi.org/10.3390/molecules26061685 https://www.mdpi.com/journal/molecules Molecules 2021, 26, x FOR PEER REVIEW 2 of 28

Molecules 2021, 26, 1685 2 of 27

combine catalytic amounts of a dye sensitizer with visible light and yield a variety of ox- ygenatedamounts products of a dye [50–52]. sensitizer Phot withooxygenations visible light are and also yield performe a varietyd industrially of oxygenated to produce prod- low-volumeucts [50–52]. fine Photooxygenations chemicals such as are fragrances, also performed flavors industrially or pharmaceuticals to produce [53–55]. low-volume These featuresﬁne chemicals make photooxygenations such as fragrances, ﬂavorsideal candid or pharmaceuticalsates for solar chemical [53–55]. Thesemanufacturing features make processes.photooxygenations ideal candidates for solar chemical manufacturing processes.

2.2. Solar Solar Reactors Reactors for for Synthetic Synthetic Applications Applications SolarSolar reactors reactors are are classified classified based based on on their their sun sun concentration factor factor (CF) (CF) and and range fromfrom non- non- to to highly highly concentrating concentrating systems. systems. Non- Non- and and low low concentrating concentrating reactors reactors are are char- char- acterizedacterized by by large large apertures apertures that that enable enable them them to to harvest harvest direct direct and and diffuse diffuse radiation. The The simplestsimplest non-concentrating non-concentrating devices areare flatbedflatbed reactors reactors with with large large surface surface areas, areas, thin thin bodies, bodies,and and operation operation volumes volumes of up of to up 30 to L (Figure30 L (Figure1a) [ 56 1a)]. These [56]. fixedThese reactors fixed reactors are best are tilted best to tiltedthe latitude to the latitude of the location of the location for optimal for optimal harvest harvest of sunlight. of sunlight. Improved Improved models models operate opin eratecirculation in circulation mode with mode an externalwith an heatexternal exchanger. heat exchanger. The compound The compound parabolic collectorparabolic (CPC) col- lectoruses a(CPC) ‘round uses W’-shaped a ‘round polished W’-shaped aluminum polished reflector aluminum that reflector is tilted tothat the is localtilted latitude to the localand directslatitude all and available directs radiationall available onto radiation a central onto received a central tube received(Figure1 b)tube [57 (Figure,58] . These 1b) [57,58].advanced These systems advanced have systems a concentration have a concentr factoration of one factor sun of (or one slightly sun (or above) slightly and above) most andcommonly most commonly operate in operate circulating in circulating batch mode batch on mode volumes on volumes of up to 100of up L. to Several 100 L. reactorsSeveral reactorscan also can be connectedalso be connected in series in to series form to larger form units. larger Parabolic units. Parabolic trough reactorstrough reactors utilize large uti- lizereflectors large reflectors that focus that only focus direct only sunlight direct sunlight onto a reaction onto a reaction tube in their tube focalin their line focal and line can andreach can concentration reach concentration factors offactors 20–150 of suns20–150 [32 su].ns More [32]. modern More modern reactors reactors such as such the former as the formerPROPHIS PROPHIS loop (PaRabolic loop (PaRabolic trough-facility trough-facility for Organic for PHotochemicalOrganic PHotochemical synthesis, synthesis, Figure1c) Figureoperate 1c) on operate volumes on up volumes to 150 Lup and to 150 track L theand movement track the movement of the sun, of either the sun, horizontally either hor- or izontallythree-dimensionally or three-dimensionally (horizontally (horizontally and vertically) and [ 59vertically)]. [59].

(a) (b) (c)

FigureFigure 1. 1. ((aa)) Circulating Circulating flatbed flatbed reactor reactor with with refl reflectiveective back back and and external external cooling cooling at at James James Cook Cook University University in in Townsville, Townsville, 2 Australia.Australia. ( (bb)) CPC CPC reactor reactor with with a a 2 2 m m2 apertureaperture at at James James Cook Cook University University in in Townsville, Townsville, Australia. ( (cc)) Former Former PROPHIS PROPHIS looploop at at the the German German Aerospace Aerospace Center Center (DLR) (DLR) in Cologne-Porz, Cologne-Porz, Germany (modified (modified from [[60]).60]).

WhileWhile solar solar dish dish concentrators concentrators can can provide provide very very high high concentration concentration factors factors of of up up to to 50005000 suns, suns, these these sophisticated sophisticated devices devices are are rare rarelyly used used for for preparative preparative solar solar photochemical photochemical applications.applications. Since thethe receiver receiver in in the the focal foca pointl point easily easily reaches reaches extreme extreme temperature temperature ranges, ranges,dish reactors dish reactors are instead are advantageousinstead advantageo for thermochemicalus for thermochemical processes processes [33]. Likewise, [33]. Like- solar wise,furnaces solar can furnaces reach extreme can reach concentration extreme concen factorstration of 5000–20,000 factors of suns 5000–20,000 by reﬂecting suns sunlight by re- flectingvia a heliostat sunlight or via heliostat a heliosta ﬁeldt onto or heliostat a concentrator, field onto which a concentrator, then focuses which the sun’s then beam focuses onto thethe sun’s experimental beam onto setup. the Theexperimental naturally extremesetup. The heat naturally generation extreme and theheat costly generation installation and thehave costly limited installation the application have limited of solar the furnaces application mainly of to materialsolar furnaces sciences mainly and solar to material thermal sciencesresearch and activities solar thermal [61–63]. research activities [61–63].

3. Photooxygenations in Organic Synthesis Photooxygenations, also known as Type II photosensitized oxidations [64], utilize 1 3 singlet oxygen ( O2), which is generated from ground-state triplet oxygen ( O2) via photo- sensitization (Scheme1)[ 65]. The sensitizer (Sens) initially absorbs light and undergoes intersystem crossing (ISC) to its triplet excited state (3Sens*). Subsequent energy transfer

Molecules 2021, 26, x FOR PEER REVIEW 3 of 28

3. Photooxygenations in Organic Synthesis Photooxygenations, also known as Type II photosensitized oxidations [64], utilize Molecules 2021, 26, 1685 3 of 27 singlet oxygen (1O2), which is generated from ground-state triplet oxygen (3O2) via photo- sensitization (Scheme 1) [65]. The sensitizer (Sens) initially absorbs light and undergoes intersystem crossing (ISC) to its triplet excited state (3Sens*). Subsequent energy transfer toto groundground statestate oxygenoxygen producesproduces singletsinglet oxygen.oxygen. TheThe dissolveddissolved oxygenoxygen concentrationconcentration inin mostmost organicorganic solventssolvents isis lowlow [[66,67]66,67] andand thus,thus, finefine streamsstreams ofof oxygenoxygen oror compressedcompressed airair areare constantlyconstantly passedpassed throughthrough thethe reactionreaction media.media. Likewise,Likewise, thethe lifetimelifetime ofof singletsinglet oxygenoxygen criticallycritically dependsdepends on on the the reaction reaction medium medium and and is highest is highest in halogenated in halogenated solvents solvents [67]. Due[67]. toDue the to environmental the environmental hazards hazards linked linked to these to th solvents,ese solvents, many many industrial industrial and solar and photooxy-solar pho- genationtooxygenation processes processes are conducted are conducted in less in harmful less harmful alcoholic alcoholic solvents solvents instead instead [68]. A [68]. series A ofseries photosensitizers of photosensitizers with large with absorption large absorption coefficients coefficients within the within visible the spectrum, visible spectrum, suitable solubilitiessuitable solubilities in a range in ofa range organic of solvents,organic solv longents, triplet long lifetimes triplet lifetimes and high and quantum high quantum yields 1 1 foryieldsO2 forgeneration O2 generation have now have become now become available availa [69,70ble]. The[69,70]. most The common most photosensitizerscommon photo- methylenesensitizers blue,methylene tetraphenylporphyrin blue, tetraphenylporphyr and rosein bengal and rose (as bengal disodium (as salt)disodium are shown salt) are in Schemeshown in1. ForScheme easy 1. recovery For easy and recovery reuse, and photosensitizers reuse, photosensitizers have also beenhave immobilizedalso been immo- on anbilized inert on solid an inert support solid [71 support,72]. Photooxygenations [71,72]. Photooxygenations have furthermore have furthermore been successfully been suc- performedcessfully performed under continuous-flow under continuous-flow conditions conditions [73]. Thermal [73]. Thermal oxygenation oxygenation processes pro- in- 1 volvingcesses involvingO2 have 1O likewise2 have likewise been developed been developed but are atbut present are at lesspresent important less important in organic in synthesisorganic synthesis [74]. [74].

SchemeScheme 1.1. PhotosensitizedPhotosensitized generationgeneration ofof singletsinglet oxygenoxygen andand commoncommon photosensitizers.photosensitizers.

SingletSinglet oxygenoxygen cancan react react with with a a variety variety of of functional functional groups groups (Scheme (Scheme2)[ 2)50 [50–52].–52]. Conju- Con- gatedjugated dienes dienes preferably preferably undergo undergo [4+2] [4+2] cycloadditions cycloadditions to endoperoxides, to endoperoxides, while while electron-rich electron- oleﬁnsrich olefins favor favor [2+2] [2+2] cycloadditions cycloadditions to 1,2-dioxetanes. to 1,2-dioxetanes. Inactivated Inactivated oleﬁns olefins with allylic with hydro-allylic genhydrogen atoms atoms undergo undergo Schenck- Schenck-ene-typeene reactions-type reactions to allylic to hydroperoxidesallylic hydroperoxides instead [instead75,76]. Molecules 2021, 26, x FOR PEER REVIEW Thioethers[75,76]. Thioethers and phosphines and phosphines produce produce the corresponding the corresponding sulfoxides sulfoxides and phosphine and phosphine oxides. 4 of 28 Alloxides. reaction All reaction types incorporate types incorporate molecular molecular oxygen oxygen into ainto molecular a molecular entity entity with with high high to quantitativeto quantitative atom atom economies. economies.

SchemeScheme 2. General 2. General photooxygenation photooxygenation reactions. reactions.

Many of the industrially relevant photochemical processes convert naturally occurring or semi-synthetic starting materials to their corresponding oxygenated products [53– 55]. Of these, essential oil derived starting materials are especially common [55,77], which makes photooxygenations promising applications for product diversification and value- adding in the essential oil industry [78]. This review will focus on demonstration (multigram)- to technical (multikilogram)-scale transformations conducted in purpose-designed solar reactors. Whenever available, key technical features of the solar reactor systems used have been provided. Related demonstration-scale photooxidations involving the super oxide anion radical (O2•–) or a thermal oxidation step were also included. In many cases, solar transformations were monitored, and obtained product mixtures analyzed by GC or GC–MS analysis techniques. Whenever necessary, the reported yields or conversions from these studies were rounded for consistency and to account for experimental errors.

4. Solar Preparative Photooxygenations 4.1. Photooxygenation of α-Terpinene and Related Reactions The first improvised solar manufacturing plant was built after the Second World War in Germany by the late Prof. Günther Otto Schenck (Figure 2). The plant produced ascaridole (2) (Scheme 3), which was urgently needed at that time to fight ascaris infections in humans [79–81]. The plant consisted of 200 carboy glass bottles of 10 L each that were placed in rows on wooden racks. Each flask was filled with 5 L of air-saturated ethanol solutions of α-terpinene (1) [54]. Stinging nettles (or spinach) leaves served as chlorophyll and hence photosensitizer source. During solar exposure, the flasks were sprayed from above with water for cooling and shaken regularly for mixing and to maintain oxygen saturation. In just two sunny summer days, the entire plant was able to produce ca. 2 kg of ascaridole (2). The application of ascaridole as an anthelmintic drug was later discouraged due to its severe side effects and its solar production was subsequently discontinued. Despite this, the photooxygenation of α-terpinene (1) still serves as a common model reaction in the development of new photosensitizers or photochemical reactor systems.

Scheme 3. General photooxygenation reactions.

Molecules 2021, 26, x FOR PEER REVIEW 4 of 28

Molecules 2021, 26, 1685 4 of 27

Many of the industrially relevant photochemical processes convert naturally occurring or semi-synthetic starting materials to their corresponding oxygenated products [53–55]. Of Scheme 2. Generalthese, essential photooxygenation oil derived starting reactions. materials are especially common [55,77], which makes photooxygenations promising applications for product diversification and value-adding in the essential oil industry [78]. This review will focus on demonstration (multigram)- Many of the industriallyto technical (multikilogram)-scale relevant photochemical transformations processes conducted convert in purpose-designed naturally occur- solar ring or semi-syntheticreactors. starting Whenever materials available, to their key technical corresponding features of oxygenated the solar reactor products systems used [53– have been provided. Related demonstration-scale photooxidations involving the super 55]. Of these, essential oil derived starting•– materials are especially common [55,77], which oxide anion radical (O2 ) or a thermal oxidation step were also included. In many cases, makes photooxygenationssolar transformations promising were applicatio monitored,ns and for obtained product product diversification mixtures analyzed and by value- GC or adding in the essentialGC–MS oil analysisindustry techniques. [78]. This Whenever review necessary, will focus the reported on demonstration yields or conversions (multi- from gram)- to technical (multikilogram)-scalethese studies were rounded fortran consistencysformations and to conducted account for experimental in purpose-designed errors. solar reactors. Whenever4. Solar available, Preparative key Photooxygenations technical features of the solar reactor systems used have been provided.4.1. Related Photooxygenation demonstration-scale of α-Terpinene and Related photooxidations Reactions involving the super The first improvised solar manufacturing plant was built after the Second World oxide anion radical (O2•–) or a thermal oxidation step were also included. In many cases, War in Germany by the late Prof. Günther Otto Schenck (Figure2). The plant produced solar transformationsascaridole were monitored, (2) (Scheme3), whichand obta was urgentlyined product needed at mixtures that time to analyzed fight ascaris infectionsby GC or GC–MS analysis techniques.in humans [Whenever79–81]. The plant necessary, consisted the of 200 reported carboy glass yields bottles or of conversions 10 L each that werefrom these studies were roundedplaced in rows for onconsistency wooden racks. and Each to flaskaccount was filled for withexperimental 5 L of air-saturated errors. ethanol solutions of α-terpinene (1)[54]. Stinging nettles (or spinach) leaves served as chlorophyll and hence photosensitizer source. During solar exposure, the flasks were sprayed from 4. Solar Preparativeabove Photooxygenations with water for cooling and shaken regularly for mixing and to maintain oxygen saturation. In just two sunny summer days, the entire plant was able to produce ca. 2 kg of 4.1. Photooxygenationascaridole of α-Terpinene (2). The application and Related of ascaridole Reactions as an anthelmintic drug was later discouraged The first improviseddue to itssolar severe manufacturing side effects and pl itsant solar was production built after was subsequentlythe Second discontinued.World War Molecules 2021, 26, x FOR PEER REVIEWDespite this, the photooxygenation of α-terpinene (1) still serves as a common5 modelof 28

in Germany by the reactionlate Prof. in the Günther development Otto of newSchenck photosensitizers (Figure or2). photochemical The plant produced reactor systems. ascaridole (2) (Scheme 3), which was urgently needed at that time to fight ascaris infections in humans [79–81]. The plant consisted of 200 carboy glass bottles of 10 L each that were placed in rows on wooden racks. Each flask was filled with 5 L of air-saturated ethanol solutions of α-terpinene (1) [54]. Stinging nettles (or spinach) leaves served as chlorophyll and hence photosensitizer source. During solar exposure, the flasks were sprayed from above with water for cooling and shaken regularly for mixing and to maintain oxygen saturation. In just two sunny summer days, the entire plant was able to produce ca. 2 kg of ascaridole (2). The application of ascaridole as an anthelmintic drug was later discour-

aged due to its severe(a) side effects and its solar production was subs(b) equently discontinued.

FigureDespiteFigure 2. 2.SolarSolar this, production production the photooxygenation plant plant for for the the synthesis synthesis of ofofascaridole ascaridole α-terpinene (2). (2 ().a) (aProf.) Prof.(1 Günther) still Günther servesOtto Otto Schenck Schenck as a in common infront front of ofa row a model row of of re- reactionactionreaction bottles bottlesin the(taken (taken development from from [47]). [47 ]).(b) ( bThe) Theof entire entirenew plant plant photosensi in inZiegelhausen, Ziegelhausen,tizers Ge Germany,rmany, or photochemi around around 1949 1949 (taken (takencal fromreactor from [54]). [54]). systems.

The photooxidation of α-terpinene (1) in the presence of a benzocoumarin sensitizer (3) and in concentrated sunlight has been reported by Karapire and co-workers (Scheme 4) [82]. The authors used a small Fix Focus dish reactor with a cylindrical water-cooled Pyrex photoreactor vessel in its focal point (Figure 3). The dish concentrator had a focal length of 0.65 m, a usable reflective surface area of 2.66 m2 (total of 3.68 m2) and a reflective aluminized polymer film cover (40 μm thick layer with 0.91 g/cm3 of alumina). The device tracked the sun only for the elevation and required manual focusing for its azimuth. The concentration factor at the reactor’s focal point could be adjusted between 40 and 150 suns dependingScheme on experimental 3. General photooxygenationrequirements. Solutions reactions. of 1 (0.1 M) and catalytic amounts of Scheme 3. General photooxygenation reactions. 3 in acetonitrile were exposed to concentrated sunlight (at 130–150 suns) for 15 min and 4 h while air was passed through the reaction mixtures. Portions of the solutions were subsequently treated with sodium sulfite (Na2SO3) and analyzed by GC–MS. Photoproducts were assigned based on their MS-spectra and characteristic molecular ions but were not isolated. Short illumination for 15 min resulted in an incomplete conversion of 1 of approx. 83% and furnished p-cymene (4) as the main product in ca. 49%. In addition, ascaridole (2), and the ring-opened compounds 5 and 6 were identified by GC–MS in smaller amounts of approx. 8%, 6% and 1%, respectively. In contrast, prolonged exposure to concentrated sunlight for 4 h gave complete conversion of 1 and produced p-cymene (4) and ascaridole (2) as the main products in amounts of ca. 40% and 27%. The ring-opened products 5, 6 and 7 were additionally formed with approx. 7%, 3% and 2%, respectively. When the solar illumination for 4 h was repeated with nitrogen purging, no products were obtained. Based on these findings, the authors suggested that photooxidations involved both singlet oxygen (1O2) as well as super oxide anion radical (O2•–) pathways.

Scheme 4. Photooxidation of α-terpinene (1) with benzocoumarin sensitizer 3.

Molecules 2021, 26, x FOR PEER REVIEW 5 of 28

(a) (b) Figure 2. Solar production plant for the synthesis of ascaridole (2). (a) Prof. Günther Otto Schenck in front of a row of reaction bottles (taken from [47]). (b) The entire plant in Ziegelhausen, Germany, around 1949 (taken from [54]).

Molecules 2021, 26The, 1685 photooxidation of α-terpinene (1) in the presence of a benzocoumarin sensitizer 5 of 27 (3) and in concentrated sunlight has been reported by Karapire and co-workers (Scheme 4) [82]. The authors used a small Fix Focus dish reactor with a cylindrical water-cooled Pyrex photoreactor vessel in its focal point (Figure 3). The dish concentrator had a focal The photooxidation of α-terpinene (1) in the presence of a benzocoumarin sensitizer (3) 2 2 length of 0.65 m, a usableand in concentrated reflective surface sunlight area has been of 2.66 reported m (total by Karapire of 3.68 and m co-workers) and a reflective(Scheme 4)[82] . aluminized polymerThe film authors cover used (40 aμ smallm thick Fix layer Focus with dish 0.91 reactor g/cm with3 of a alumina). cylindrical The water-cooled device Pyrex tracked the sun onlyphotoreactor for the elevation vessel in and its focal requ pointired (Figure manual3). Thefocusing dish concentrator for its azimuth. had a focalThe length concentration factorof at 0.65 the m, reactor’s a usable focal reflective point surface could area be adjusted of 2.66 m between2 (total of 40 3.68 and m2 150) and suns a reflective 3 depending on experimentalaluminized requirements. polymer film cover Solutions (40 µm of thick 1 (0.1 layer M) and with catalytic 0.91 g/cm amountsof alumina). of The device tracked the sun only for the elevation and required manual focusing for its azimuth. 3 in acetonitrile were exposed to concentrated sunlight (at 130–150 suns) for 15 min and 4 The concentration factor at the reactor’s focal point could be adjusted between 40 and h while air was passed150 suns through depending the reaction on experimental mixtures. requirements. Portions of Solutions the solutions of 1 (0.1 were M) andsub- catalytic sequently treated withamounts sodium of 3 insulfite acetonitrile (Na2SO were3) and exposed analyzed to concentrated by GC–MS. sunlight Photoproducts (at 130–150 suns) were assigned basedfor on 15 mintheir and MS-spectra 4 h while air and was characteristic passed through molecular the reaction ions mixtures. but were Portions not of the isolated. Short illuminationsolutions were for 15 subsequently min resulted treated in an with incomplete sodium sulfite conversion (Na2SO3) andof 1 analyzed of approx. by GC–MS. 83% and furnishedPhotoproducts p-cymene (4 were) as the assigned main based product on their in MS-spectraca. 49%. In and addition, characteristic ascaridole molecular ions but were not isolated. Short illumination for 15 min resulted in an incomplete conversion (2), and the ring-openedof 1 of approx. compounds 83% and furnished5 and 6p -cymenewere identified (4) as the main by productGC–MS in ca.in 49%.smaller In addition, amounts of approx.ascaridole 8%, 6% and (2), and1%, therespective ring-openedly. In compounds contrast, prolonged5 and 6 were exposure identified to by con- GC–MS in centrated sunlight smallerfor 4 h amountsgave complete of approx. conversion 8%, 6% and of 1%, 1 respectively.and produced In contrast, p-cymene prolonged (4) and exposure ascaridole (2) as theto main concentrated products sunlight in amounts for 4 h gave of ca. complete 40% and conversion 27%. The of ring-opened1 and produced prod-p-cymene (4) ucts 5, 6 and 7 wereand additionally ascaridole ( 2formed) as the mainwith products approx. in 7%, amounts 3% and of ca.2%, 40% respectively. and 27%. The When ring-opened the solar illuminationproducts for 45 h, 6 wasand 7repeatedwere additionally with nitrogen formed purging, with approx. no products 7%, 3% and were 2%, respectively.ob- When the solar illumination for 4 h was repeated with nitrogen purging, no products were tained. Based on theseobtained. findings, Based the on authors these findings, suggested the authors that photooxidations suggested that photooxidations involved both involved 1 2 1 2•– •– singlet oxygen ( Oboth) as singletwell as oxygen super( oxideO2) as anion well as radical super oxide (O anion) pathways. radical (O 2 ) pathways.

Molecules 2021, 26, x FOR PEER REVIEW 6 of 28

Scheme 4. PhotooxidationScheme 4. Photooxidation of α-terpinene of α (-terpinene1) with benzocoumarin (1) with benzocoumarin sensitizer sensitizer 3. 3.

FigureFigure 3.3. ParabolicParabolic dishdish solarsolar concentratorconcentrator (HTC(HTC SolarSolar ResearchResearch GmbH,GmbH, Germany)Germany) atat EgeEge UniversityUniver- sity in Izmir, Turkey, around 2001 (taken from [83]). in Izmir, Turkey, around 2001 (taken from [83]). In an extension of this work, the same authors utilized perylenediimides as sensitizers in solution or immobilized in PVC or as sol-gel thin films on glass slides for the photooxidation of α-terpinene (1) (Scheme 5) [84]. Illumination experiments were conducted for 15 min and 1 h in the same small parabolic dish reactor using concentration factors of 80–90 suns (Figure 3). After treatment of the reaction mixture with Na2SO3, the products obtained were detected and assigned by GC–MS analysis. The outcome of the reactions critically depended on the photosensitizer material and solar exposure conditions. In acetonitrile solution (0.1 M of 1) with trace amounts of ABIPER photosensitizer (8a), solar exposure for 1 h gave ascaridole (2) as the dominant photoproduct in 96%. Under the same conditions, the reaction with the 8a-containing sol-gel thin film remained incomplete with 20% of residual 1, and furnished ascaridole (2) and p-cymene (4) in amounts of ca. 49% and 18% as the main products. In contrast, the respective illumination with the PVC-immobilized sensitizer 8a predominantly formed 3,6-dioxoheptanal (9) in 55%, next to 21% of ascaridole (2), 9% of p-cymene (4) and 16% of terpinolene (10), respectively. With soluble and sol-gel immobilized 8a, shorter illuminations in concentrated sunlight for just 15 min resulted in more complex product mixtures. When N-DODEPER (8b) was used as photosensitizer in acetonitrile solution, solar exposure for 1 h resulted in approx. 70% conversion and yielded predominantly p-cymene (4) and 3-isopropyl-2-heptanal-6-one (5) with ca. 51% and 10%, respectively. After the same illumination time, the corresponding 8b sol-gel thin film produced a conversion of 80% and furnished p-cymene (4) and ascaridole (2) as major products in ca. 34% and 15%, as determined by GC–MS. The corresponding solar reaction with 8b-impregnated PVC generated ca. 42% of 3,6-dioxoheptanal (9), 30% of p-cymene (4), 11% of terpinolene (10) and 9% of ascaridole (2) as the main photoproducts instead. Shorter illuminations for 15 min again delivered different product mixtures and compositions. In the absence of any photosensitizer, GC–MS analysis after solar exposure for 1 h revealed an incomplete conversion of 1 of 57% with p-cymene (4), 3,6-dioxoheptanal (9), ascaridole (2), terpinolene (10) and cineol (11) as products in ca. 17%, 16%, 10%, 9% and 6%. In contrast, exposure to concentrated sunlight for just 15 min gave a very incomplete reaction (<7% conversion) with cineole (11) as the only product in approx. 7%. The formation of terpinolene (10) was suggested to occur via thermal rearrangement of 1. For all other products, the authors proposed reaction pathways via the super oxide anion radical (O2•–). The variable product compositions and the small scales (25 mL) prevent any wider exploitation of this synthesis methodology.

Molecules 2021, 26, 1685 6 of 27

In an extension of this work, the same authors utilized perylenediimides as sensitizers in solution or immobilized in PVC or as sol-gel thin films on glass slides for the photooxidation of α-terpinene (1) (Scheme5)[ 84]. Illumination experiments were conducted for 15 min and 1 h in the same small parabolic dish reactor using concentration factors of 80–90 suns (Figure3). After treatment of the reaction mixture with Na 2SO3, the products obtained were detected and assigned by GC–MS analysis. The outcome of the reactions critically depended on the photosensitizer material and solar exposure conditions. In acetonitrile solution (0.1 M of 1) with trace amounts of ABIPER photosensitizer (8a), solar exposure for 1 h gave ascaridole (2) as the dominant photoproduct in 96%. Under the same conditions, the reaction with the 8a-containing sol-gel thin film remained incomplete with 20% of residual 1, and furnished ascaridole (2) and p-cymene (4) in amounts of ca. 49% and 18% as the main products. In contrast, the respective illumination with the PVC-immobilized sensitizer 8a predominantly formed 3,6-dioxoheptanal (9) in 55%, next to 21% of ascaridole (2), 9% of p-cymene (4) and 16% of terpinolene (10), respectively. With soluble and sol-gel immobilized 8a, shorter illuminations in concentrated sunlight for just 15 min resulted in more complex product mixtures. When N-DODEPER (8b) was used as photosensitizer in acetonitrile solution, solar exposure for 1 h resulted in approx. 70% conversion and yielded predominantly p-cymene (4) and 3-isopropyl-2-heptanal-6-one (5) with ca. 51% and 10%, respectively. After the same illumination time, the corresponding 8b sol-gel thin film produced a conversion of 80% and furnished p-cymene (4) and ascaridole (2) as major products in ca. 34% and 15%, as determined by GC–MS. The corresponding solar reaction with 8b-impregnated PVC generated ca. 42% of 3,6-dioxoheptanal (9), 30% of p-cymene (4), 11% of terpinolene (10) and 9% of ascaridole (2) as the main photoproducts instead. Shorter illuminations for 15 min again delivered different product mixtures and compositions. In the absence of any photosensitizer, GC–MS analysis after solar exposure for 1 h revealed an incomplete conversion of 1 of 57% with p-cymene (4), 3,6-dioxoheptanal (9), ascaridole (2), terpinolene (10) and cineol (11) as products in ca. 17%, 16%, 10%, 9% and 6%. In contrast, exposure to concentrated sunlight for just 15 min gave a very incomplete reaction (<7% conversion) with cineole (11) as the only product in approx. 7%. The formation of terpinolene (10) was suggested to occur via thermal rearrangement of 1. For all other products, the •– authors proposed reaction pathways via the super oxide anion radical (O2 ). The variable Molecules 2021, 26, x FOR PEER REVIEW 7 of 28 product compositions and the small scales (25 mL) prevent any wider exploitation of this synthesis methodology.

SchemeScheme 5. PerylenediimidePerylenediimide sensitizers 8a8a andand bb usedused andandadditional additional photoproducts photoproducts9–11 9-11obtained obtained during during solar solar photooxida- photooxi- dationstions of ofα-terpinene. α-terpinene.

TheThe photodehydrogenation photodehydrogenation of of αα-terpinene-terpinene ( (11)) to to pp-cymene-cymene ( (44)) in in concentrated concentrated sun- sun- lightlight has has likewise likewise been been reported reported by by Avcibasi Avcibasi and and co-workers co-workers [85]. [85 The]. The authors authors reused reused the smallthe small Fix Focus Fix Focus dish reactor dish reactor shown shown in Figure in Figure3 but with3 but a withconcentration a concentration factor of factor 40 suns. of An40 suns.initially An degassed initially degassedsolution of solution 1, benzophenone of 1, benzophenone (Ph2CO) (Phand2 CO)cupric and pivalate cupric (CuPiv pivalate2; t ((CuPivtBuCO2);(2Cu)BuCO in carbon2)2Cu) tetrachloride in carbon tetrachloride was exposed was to exposed concentrated to concentrated sunlight over sunlight a period over ofa period4 h (Scheme of 4 h (Scheme6). During6). Duringthis exposure, this exposure, the reaction the reaction was stopped was stopped eight times eight timesfor aera- for + 2+ tion/degassingaeration/degassing that oxidized that oxidized cuprous cuprous (Cu+)(Cu back) backto cupric to cupric (Cu2+ (Cu) ions.) The ions. reaction The reaction mix- turemixture was wassubsequently subsequently analyzed analyzed by GC–MS, by GC–MS, which which revealed revealed a conversion a conversion of approx. of approx. 73% with73% withca. 47% ca. of 47% p-cymene of p-cymene (4). The (4). postulated The postulated reaction reaction mechanism mechanism involved involved hydrogen hydrogen ab- α stractionabstraction from from α-terpinene-terpinene (1) ( 1by) by triplet-excited triplet-excited benzophenone benzophenone followed followed by by reduction reduction of of 2+ thethe resulting resulting radical radical intermediate intermediate by by Cu Cu2+. .

Scheme 6. Photodehydrogenation of α-terpinene (1) to p-cymene (4).

4.2. Photooxygenation of Citronellol for the Production of Rose Oxide The Schenck ene-photooxygenation of enantiomerically pure or racemic citronellol (12) is a key step in the synthesis of rose oxide (15), which serves as an important fragrance material in perfumery (Scheme 7) [86]. The initially obtained regioisomeric hydroperoxides 13a and b are reduced to their corresponding alcohols 14a and b in reported isolated yields of 60% and 35%, respectively. Subsequent acid-catalyzed rearrangement and cyclization of 14a gives rose oxide (15) as a mixture of cis- and trans-diastereoisomers. In- dustrially, the photooxygenation of 12 is typically performed in methanol using rose bengal (RB) as a photosensitizer. After the additional thermal steps, rose oxide 15 is isolated from the crude reaction mixture by steam distillation in an overall yield of 50–60%. The annual industrial production of rose oxide has been estimated to >100 tons [87]. A modified two-phase liquid/liquid cyclization process that utilizes both regioisomers 14a and b has also been patented [88].

Molecules 2021, 26, x FOR PEER REVIEW 7 of 28

Scheme 5. Perylenediimide sensitizers 8a and b used and additional photoproducts 9-11 obtained during solar photooxidations of α-terpinene.

The photodehydrogenation of α-terpinene (1) to p-cymene (4) in concentrated sunlight has likewise been reported by Avcibasi and co-workers [85]. The authors reused the small Fix Focus dish reactor shown in Figure 3 but with a concentration factor of 40 suns. An initially degassed solution of 1, benzophenone (Ph2CO) and cupric pivalate (CuPiv2; (tBuCO2)2Cu) in carbon tetrachloride was exposed to concentrated sunlight over a period of 4 h (Scheme 6). During this exposure, the reaction was stopped eight times for aeration/degassing that oxidized cuprous (Cu+) back to cupric (Cu2+) ions. The reaction mixture was subsequently analyzed by GC–MS, which revealed a conversion of approx. 73% with ca. 47% of p-cymene (4). The postulated reaction mechanism involved hydrogen ab- Molecules 2021straction, 26, 1685 from α-terpinene (1) by triplet-excited benzophenone followed by reduction 7of of 27 the resulting radical intermediate by Cu2+.

Scheme 6. Photodehydrogenation of α-terpinene (1) to p-cymene (4). Scheme 6. Photodehydrogenation of α-terpinene (1) to p-cymene (4). 4.2. Photooxygenation of Citronellol for the Production of Rose Oxide 4.2. Photooxygenation Theof Citronello Schenck enel for-photooxygenation the Production of of Rose enantiomerically Oxide pure or racemic citronellol The Schenck(12 ene) is-photooxygenation a key step in the synthesis of ofenantiomerical rose oxide (15), whichly pure serves or asracemic an important citronellol fragrance material in perfumery (Scheme7)[ 86]. The initially obtained regioisomeric hydroperoxides (12) is a key step in13a theand synthesisb are reduced of rose to their oxide corresponding (15), which alcohols serves14a asand anb importantin reported isolatedfragrance yields material in perfumeryof 60% and(Scheme 35%, respectively. 7) [86]. The Subsequent initially acid-catalyzedobtained regioisomeric rearrangement hydroperox- and cyclization ides 13a and b areof reduced14a gives to rose their oxide corresponding (15) as a mixture alcohols of cis- and 14atrans and-diastereoisomers. b in reported Industrially,isolated yields of 60% andthe 35%, photooxygenation respectively. of 12Subsequentis typically performed acid-catalyzed in methanol rearrangement using rose bengal and (RB) cy- as a photosensitizer. After the additional thermal steps, rose oxide 15 is isolated from the crude clization of 14a givesreaction rose mixture oxide by ( steam15) as distillation a mixture in anof overall cis- and yield trans of 50–60%.-diastereoisomers. The annual industrial In- dustrially, the phproductionotooxygenation of rose oxideof 12 hasis typically been estimated performed to >100 in tons methanol [87]. A modiﬁedusing rose two-phase ben- Molecules 2021, 26, x FOR PEER REVIEW 8 of 28 gal (RB) as a photosensitizer.liquid/liquid cyclization After the process additional that utilizes thermal both steps, regioisomers rose oxide14a and 15b ishas isolated also been from the crude reactionpatented [mixture88]. by steam distillation in an overall yield of 50–60%. The annual industrial production of rose oxide has been estimated to >100 tons [87]. A modified two-phase liquid/liquid cyclization process that utilizes both regioisomers 14a and b has also been patented [88].

SchemeScheme 7. Photooxygenation 7. Photooxygenation of of citronellol citronellol ( (12) and subsequentsubsequent conversion conversion to roseto rose oxide oxide (15). (15). Demuth and co-workers have developed simple flatbed reactors for solar photochemi- calDemuth and -thermal and operationsco-workers in countrieshave developed with largely simple diffuse flatbed solar radiation reactors [89 ,for90]. solar Several photo- chemicalmodels and of these -thermal low-tech/low-price operations in Universal countries Light with Collector largely diffuse and Reactors solar (ULCR)radiation were [89,90]. Severalconstructed models and of subsequently these low-tech/low-price tested for a range Universal of photochemical Light transformations.Collector and TheReactors (ULCR)simple were ULCR-1T constructed model (Figure and subsequently4a) was built te fromsted an for aluminum a range of frame photochemical and a reflective transformations.stainless-steel The simple bottom ULCR-1T plate with model a Teflon (Figure® cover 4a) foil was (transparency built from >300 an aluminum nm). The reactor frame and a reflectivewas placed stainless-steel horizontally onbottom the ground plate andwith carried a Teflon inlet® cover and outlet foil (transparency ports on the side >300 of nm). Thethe reactor frame, was a gas placed inlet, ahorizontally pressure release on the valve ground and optionaland carried water inlet cooling. and outlet The more ports on advanced and larger ULCR-4P (Figure4b) was made from Plexiglas ® (poly(methyl 2- the side of the frame, a gas inlet, a pressure release valve and optional water cooling. The methylpropenoate), PMMA), which has a good light transparency of >300 nm. The reactor ® morehad advanced an aperture and of 1.5 larger m2 (98 ULCR-4P W × 165 H(Figure cm), a thickness4b) was made of 1.5 cmfrom and Plexiglas a volume of (poly(methyl approx. 2-methylpropenoate),20 L. An inlet port for gasesPMMA), and additionalwhich has inlet a good and outlet light portstransparency for reagents of were >300 attached nm. The re- actorat thehad top an of aperture the reactor. of 1.5 The m gas-feeding2 (98 W × 165 tube H ran cm), down a thickness the inside of of 1.5 the cm reactor’s and a volume body of approx.with a 20 perforated L. An inlet end port along for its gases bottom. and During additi operation,onal inlet the and whole outlet device ports was for positioned reagents were attachedin the north–southat the top of direction the reactor. and The tilted gas-feed to approximatelying tube ran 45◦ down. The alternative the inside ULCR-4Mof the reactor’s bodymodel with (not a shown)perforated was end made along from moreits bottom. chemically During inert operation, twin-walled the Makrolon whole® device(poly- was positioned in the north–south direction and tilted to approximately 45°. The alternative ULCR-4M model (not shown) was made from more chemically inert twin-walled Mak- rolon® (polycarbonate, PC). This material shows a lower light transparency of >400 nm, which makes it beneficial for solar transformations that do not require or are sensitive to the UV fraction of sunlight.

(a) (b) Figure 4. Examples of flatbed reactors developed by Prof Martin Demuth at the former Max Planck Institute for Radiation Chemistry in Mühlheim/Ruhr, Germany, between 1998–2000 (taken from [90]). (a) Horizonal ULCR-1T reactors. (b) Tilted ULCR-4P reactors.

Selected reactor models were subsequently used for the rose bengal-sensitized photooxygenation of citronellol (12) in methanol (Table 1). Using the horizontal ULCR-1T reactor, solar exposure of a 6.75 M solution of 12 for 3 days under partially cloudy conditions, followed by thermal conversions, furnished rose oxide (15) in an isolated yield of

Molecules 2021, 26, x FOR PEER REVIEW 8 of 28

Scheme 7. Photooxygenation of citronellol (12) and subsequent conversion to rose oxide (15).

Demuth and co-workers have developed simple flatbed reactors for solar photochemical and -thermal operations in countries with largely diffuse solar radiation [89,90]. Several models of these low-tech/low-price Universal Light Collector and Reactors (ULCR) were constructed and subsequently tested for a range of photochemical transformations. The simple ULCR-1T model (Figure 4a) was built from an aluminum frame and a reflective stainless-steel bottom plate with a Teflon® cover foil (transparency >300 nm). The reactor was placed horizontally on the ground and carried inlet and outlet ports on the side of the frame, a gas inlet, a pressure release valve and optional water cooling. The more advanced and larger ULCR-4P (Figure 4b) was made from Plexiglas® (poly(methyl 2-methylpropenoate), PMMA), which has a good light transparency of >300 nm. The reactor had an aperture of 1.5 m2 (98 W × 165 H cm), a thickness of 1.5 cm and a volume of approx. 20 L. An inlet port for gases and additional inlet and outlet ports for reagents were attached at the top of the reactor. The gas-feeding tube ran down the inside of the reactor’s Molecules 2021, 26, 1685 body with a perforated end along its bottom. During operation, the whole device8 ofwas 27 positioned in the north–south direction and tilted to approximately 45°. The alternative ULCR-4M model (not shown) was made from more chemically inert twin-walled Mak- rolon® (polycarbonate, PC). This material shows a lower light transparency of >400 nm, carbonate, PC). This material shows a lower light transparency of >400 nm, which makes it which makes it beneficial for solar transformations that do not require or are sensitive to beneﬁcial for solar transformations that do not require or are sensitive to the UV fraction the UV fraction of sunlight. of sunlight.

(a) (b)

FigureFigure 4.4. ExamplesExamples ofof ﬂatbedflatbed reactorsreactors developeddeveloped byby ProfProf MartinMartin DemuthDemuth atat thethe formerformer MaxMax PlanckPlanck InstituteInstitute forfor RadiationRadiation ChemistryChemistry inin Mühlheim/Ruhr,Mühlheim/Ruhr, Germany, betweenbetween 1998–20001998–2000 (taken(taken fromfrom [[90]).90]). ((aa)) HorizonalHorizonal ULCR-1TULCR-1T reactors.reactors. ((bb)) TiltedTilted ULCR-4PULCR-4P reactors.reactors.

SelectedSelected reactor modelsmodels were subsequentlysubsequently used for the rose bengal-sensitizedbengal-sensitized pho- tooxygenationtooxygenation of citronellol citronellol ( (1212)) in in methanol methanol (Table (Table 1).1). Using Using the the horizontal horizontal ULCR-1T ULCR-1T re- reactor,actor, solar solar exposure exposure of of a a6.75 6.75 M M solution solution of of 1212 forfor 3 3 days days under partially cloudy conditions,tions, followedfollowed byby thermalthermal conversions,conversions, furnishedfurnished roserose oxideoxide ((15)) inin anan isolatedisolated yieldyield ofof 50% (reaction 1). In a different, not specified device made from two individual reactors connected in series, circulation of a 1.1 M solution of 12 for 11 h under mainly cloudy conditions gave complete conversion and, after further treatment, rose oxide (15) was isolated in 52% yield (reaction 2). For comparison, reactions 3 and 4 in a non-specified flatbed reactor were conducted in parallel in a parabolic trough reactor of a similar size (aperture of 1 m2)[90]. Under mainly cloudy and thus diffuse radiation conditions (reaction 3), the concentrating solar trough reactor produced a poor conversion of 12 of just about 25% after 11 h of exposure, whereas the flatbed device showed near complete consumption of citronellol (12). Under sunny conditions (reaction 4), both reactor systems performed equally well and effectively reached full conversions of 12 after 8 h of illumination. The latter two experiments demonstrated the superiority of the flatbed technology under non- ideal solar conditions with predominantly diffuse radiation. In all cases, the rose oxide (15) mixtures obtained were identical in quality to those produced from reactions using artificial light.

Table 1. Experimental details of photooxygenations of citronellol (12) conducted in ﬂatbed reactors.

Reaction 1 [89] Reaction 2 [89] Reaction 3 [90] Reaction 4 [90] citronellol (12) 0.5 L (2.7 mol) 0.2 L (1.1 mol) 0.1 L (0.55 mol) 0.1 L (0.55 mol) methanol 400 mL 1 L 2 L 2 L rose bengal 1 5 × 6g 3g+1g 5g 5g reactor batch (1 m2) in series (1 m2) batch (1 m2) batch (1 m2) operation static ﬂow (2 L/h) static static time 3 days 11 h 11 h 7 h 50% cloudy, 70% cloudy; 70% cloudy; solar conditions 100% sunny 50% sunny 30% sunny 30% sunny yield of 15 2 50% (1.36 mol) 52% (0.57 mol) not reported 3 not reported 3 1 Added in several dosages. 2 Isolated yield based on 12 after further reduction and cyclization. 3 >95% conversion of 12. Molecules 2021, 26, 1685 9 of 27

Technical-scale photooxygenations of citronellol (12) have been conducted in the advanced PROPHIS loop (Figure1c). [ 59,91]. Its Helioman module comprised of four parabolic troughs equipped with Pyrex glass tubes (length 4.5 m; inner diameter 55 mm) in their focal lines. The troughs could be connected in series or in parallel. The entire concentrator had a total aperture of 32 m2 with 8 elements per trough. The individual mirror elements of 1 m2 each were made of silver-coated glass, which does not reflect any UV-light. Depending on the desired application, alternative aluminum or holographic reflectors with different reflective profiles could be fitted instead [92,93]. The reactor had a sunlight concentration factor of 30–32 suns and tracked the sun three-dimensionally with an advanced two-axis system during operation. Depending on the number of troughs employed, the operation volume ranged from 35 to 120 L. The reaction reservoir, pump, heat exchanger and gas-feeding equipment were placed in a nearby shed. Solar exposures were conducted on two different technical scales (Table2). Isopropanol was chosen as a non- hazardous solvent and oxygen gas was injected into the reagent stream, the later circulating at 30 L/min. The first experiment (reaction 5) conducted with one trough rapidly consumed 31.8 mol of 12 within three hours of solar exposure. Subsequent reduction with aqueous sodium sulfite (Na2SO3) furnished the regioisomeric diols 14a and b in a ratio of 55:45 (by GC). The second run (reaction 6) utilized all 4 troughs under optimal solar conditions and 1 almost completely converted 43.9 mol of citronellol (12) within just 2 4 h. After further thermal reduction and cyclization, rose oxide (15) was isolated in excellent quality and yield of 55%. When the PROPHIS loop was equipped with a modified receiver tube that narrowed from 5 to 4.2 cm in inner diameter, the photooxygenation of 12 proceeded somewhat faster (reaction 7) than in the experiment with an unmodified tube (reaction 8) [94]. Although the overall reaction times were similar, the reaction conducted with the modified tube consumed 12.8 mol of 12 after one hour of illumination, compared to 10.5 mol of 12 with the standard tube. This improvement was explained by a better mixing that compensated for the depletion of oxygen along the narrowing tube.

Table 2. Experimental details of photooxygenations of 12 conducted in the PROPHIS loop.

Reaction 5 [91] Reaction 6 [91] Reaction 7 [94] Reaction 8 [94] citronellol (12) 5.8 L (31.8 mol) 8 L (43.9 mol) 5.8 L (31.8 mol) 5.8 L (31.8 mol) isopropanol 40 L 72 L 40 L 40 L rose bengal 20 g 1 36 g 1 2 g 2 g O2-ﬂow 600 L/h 200 L/h not reported not reported # of troughs 1 (8 m2) 4 (32 m2) 1 (8 m2) 2 1 (8 m2) 3 1 1 1 time 3 h 2 4 h 2 /3 h 2 4 h solar conditions largely sunny sunny sunny sunny conversion of 12 4 >95% >95% (55% 5) >95% >95% 1 Added in several dosages. 2 Modiﬁed tube with narrowing diameter. 3 Time until maximum conversion was reached. 4 Determined by 5 GC (vs. internal standard) after reduction with Na2SO3. Isolated yield of 15 based on 12 after further reduction and cyclization.

To reduce heat-up and hence cooling-water needs, holographic mirrors were furthermore considered for solar photooxygenations [92,93]. These advanced reflectors were adapted to the sensitizer rose bengal with a reflectivity range of 480–620 nm. One trough of the PROPHIS loop was subsequently fitted with a holographic concentrator (Figure5a). Due to difficulties with the stability of the holographic film, no experimental results have been reported publicly. Preliminary warm-up tests were, however, successfully completed in a smaller custom-built laboratory reactor with an aperture of 0.188 m2 (20 × 94 cm) and fitted with either aluminum or holographic elements (Figure5b) [ 92]. During the experiments, a solution of rose bengal (0.4 g/L) in isopropanol was passed through the loop at a flow rate of 60 mL/min and the temperature differences between in- and outlet of the receiver tube were recorded. In comparison with the holographic mirror, the heat-up during illumination using the aluminum reflector was 60% higher due to the significant energy input of infrared radiation. Despite these encouraging results, experimental findings on the preparative photooxygenation of 12 in this device have not been published. Molecules 2021, 26, x FOR PEER REVIEW 10 of 28

to 10.5 mol of 12 with the standard tube. This improvement was explained by a better mixing that compensated for the depletion of oxygen along the narrowing tube.

Table 2. Experimental details of photooxygenations of 12 conducted in the PROPHIS loop.

reaction 5 [91] reaction 6 [91] reaction 7 [94] reaction 8 [94] citronellol (12) 5.8 L (31.8 mol) 8 L (43.9 mol) 5.8 L (31.8 mol) 5.8 L (31.8 mol) isopropanol 40 L 72 L 40 L 40 L rose bengal 20 g 1 36 g 1 2 g 2 g O2-flow 600 L/h 200 L/h not reported not reported # of troughs 1 (8 m2) 4 (32 m2) 1 (8 m2) 2 1 (8 m2) time 3 3 h 2¼ hours 2⅓ hours 2¼ hours solar conditions largely sunny sunny sunny sunny conversion of 12 4 >95% >95% (55% 5) >95% >95% 1 Added in several dosages. 2 Modified tube with narrowing diameter. 3 Time until maximum conversion was reached. 4 Determined by GC (vs. internal standard) after reduction with Na2SO3. 5 Isolated yield of 15 based on 12 after further reduction and cyclization. To reduce heat-up and hence cooling-water needs, holographic mirrors were furthermore considered for solar photooxygenations [92,93]. These advanced reflectors were adapted to the sensitizer rose bengal with a reflectivity range of 480–620 nm. One trough of the PROPHIS loop was subsequently fitted with a holographic concentrator (Figure 5a). Due to difficulties with the stability of the holographic film, no experimental results have been reported publicly. Preliminary warm-up tests were, however, successfully completed in a smaller custom-built laboratory reactor with an aperture of 0.188 m2 (20 × 94 cm) and fitted with either aluminum or holographic elements (Figure 5b) [92]. During the experiments, a solution of rose bengal (0.4 g/L) in isopropanol was passed through the loop at a flow rate of 60 mL/min and the temperature differences between in- and outlet of the receiver tube were recorded. In comparison with the holographic mirror, the heat- up during illumination using the aluminum reflector was 60% higher due to the signifi- Molecules 2021, 26, 1685 cant energy input of infrared radiation. Despite these encouraging results, experimental10 of 27 findings on the preparative photooxygenation of 12 in this device have not been published.

(a) (b)

FigureFigure 5. 5.ParabolicParabolic trough trough concentrators concentrators equi equippedpped with with holographic holographic mirror mirror elements elements at atthe the German German Aerospace Aerospace Center Center (DLR)(DLR) in inCologne-Porz, Cologne-Porz, Germany. Germany. (a ()a Former) Former PROPHIS PROPHIS loop, loop, around around 2002 2002 (taken (taken from from [39]). [39]). (b ()b Custom-built) Custom-built laboratory laboratory reactor, around 1998 (taken from [92]). reactor, around 1998 (taken from [92]).

DueDue to toits its technical technical importance importance [54,55], [54,55 ],the the photooxygenation photooxygenation of ofcitronellol citronellol (12 (12) was) was chosenchosen for for a areactor reactor comparison comparison study study [39,41]. [39,41]. Next Next to to the the advanced advanced PROPHIS PROPHIS concentratorconcentrator(Figure (Figure1c), 1c), four four additional additional non-concentrating non-concentrating reactor reactor systems systems of different of different volumes, volumes, aper- apertures,tures, concentration concentration factors, factors, operation operation modes modes and and oxygen oxygen feeding feeding systems systems were were examined examined(Table 3(Table). The 3). devices The devices were tilted were at tilted 51 ◦ butat 51° did but not did follow not thefollow movement the movement of the sun. of the The sun.CPC The system CPC system (Figure 6(Figurea) consisted 6a) consisted of 8 rows of of 8 tubesrows thatof tubes ran horizontallythat ran horizontally across three across com- threepartments compartments and were and connected were connected in series in via se U-turnries via pipes. U-turn The pipes. ﬂatbed The reactor flatbed(Figure reactor6b) was an improved ULCR-4M model constructed from a triple walled Makrolon® sheet. A cooling tube ran through its vertical chambers and a perforated gas-feeding tube along its bottom. The horizontal tube system (Figure6c) was made from 10 parallel glass tubes

connected in series via U-turn pipes. The vertical tube reactor (Figure6d) was constructed from 10 glass tubes in parallel that were fed from an inlet tube on the bottom.

Table 3. Reactor and experimental details of the solar campaign.

Reactions 9 Reactions 10 Reactions 11 Reactions 12 Reactions 13 reactor PROPHIS CPC ﬂatbed horizontal tubes vertical tubes CF 32 suns 1 sun 1 sun 1 sun 1 sun aperture 32 m2 3 m2 1.5 m2 not given not given volume 80 L (37 L 1) 40 L (26 L 1) 21 L 1 25 L (11 L 1) 20 L (10 L 1) operation circulation circulation static circulation circulation mixing static mixer 7 × 180◦ turns air bubbling 9 × 180◦ turns air bubbling citronellol (12) 8 L (43.9 mol) 4 L (22 mol) 2.1 L (11.5 mol) 2.5 L (13.7 mol) 2 L (11 mol) isopropanol 72 L 36 L 18.9 L 22.5 L 18 L rose bengal 2 40 g 20 g 10.5 g 12.5 g 10 g O2 source oxygen air air air air time 2 1 h 3/not 2 7 h/30 h 15 h/30 h 12 h/33 h 13 h/33 h (sunny/cloudy) 4 operated >95% 3/not conversion of 12 5 >95%/>95% >95%/>95% >95%/>65% >95%/>65% operated 1 Exposed volumes (all tubes). 2 Added in several dosages. 3 Reaction 6 from Table2. 4 Illumination time under sunny/cloudy conditions. 5 Determined by GC (vs. internal standard) after subsequent reduction with Na2SO3. Molecules 2021, 26, x FOR PEER REVIEW 11 of 28

(Figure 6b) was an improved ULCR-4M model constructed from a triple walled Mak- rolon® sheet. A cooling tube ran through its vertical chambers and a perforated gas-feeding tube along its bottom. The horizontal tube system (Figure 6c) was made from 10 parallel glass tubes connected in series via U-turn pipes. The vertical tube reactor (Figure 6d) Molecules 2021, 26, 1685 was constructed from 10 glass tubes in parallel that were fed from an inlet tube11 on of the 27 bottom.

(a) (b)

FigureFigure 6. 6.Solar Solar reactors reactors (except (except PROPHIS PROPHIS loop) loop) used used for for photooxygenations photooxygenations of of citronellol citronellol ( 12(12)) at at the the German German Aerospace Aerospace ® CenterCenter (DLR) (DLR) in in Cologne-Porz, Cologne-Porz, Germany Germany in in 2002. 2002. (a ()a CPC) CPC reactor. reactor. (b ()b Flatbed) Flatbed made made from from Makrolon Makrolon®.(. c()c) Horizontal Horizontal serial serial tubetube reactor. reactor. (d ()d Parallel) Parallel vertical vertical tube tube reactor. reactor.

TheThe horizontal horizontal and and vertical vertical tube tube models models had had no no sunlight sunlight concentrator concentrator and and thus thus solely solely reliedrelied on on solar solar radiation radiation hitting hitting the the surface surface of of the the glass glass tubes. tubes. Each Each reactor reactor was was loaded loaded withwith a a 10 10 vol% vol% solution solution of of12 12in in isopropanol isopropanol containing containing 0.5 0.5 g/L g/L of of rose rose bengal. bengal. Under Under ideal ideal sunnysunny conditions conditions (Table (Table3), all3), reactorsall reactors reached reached complete complete conversions conversions after 2.5after h (PROPHIS)2.5 h (PRO- toPHIS) 15 h (flatbed),to 15 h (flatbed), respectively. respectively. Due to theirDue to superior their superior oxygen oxygen feeding feeding systems systems and hence and improvedhence improved oxygen oxygen saturation, saturation, the CPC the andCPC the and horizontal the horizontal tube tube device device gave gave the the highest high- conversionest conversion based based on exposureon exposure time time and and illuminated illuminated area area values. values. The The PROPHIS PROPHIS loop loop showedshowed only only an an average average value value due due to to its its somewhat somewhat simpler simpler oxygen oxygen supply supply mechanism, mechanism, whichwhich onlyonly allowedallowed gasgas feeding at at the the bottom bottom of of the the first first trough trough with with relatively relatively large large ox- oxygenygen bubbles bubbles formed. formed. The The poorer poorer efficiencies efficiencies of the of theflatbed flatbed and and vertical vertical tube tube reactors reactors were wereexplained explained by the by inefficient the inefficient gas saturation gas saturation throughout throughout these these devices. devices. Especially Especially the com- the compartmentalizedpartmentalized structure structure of the of theflatbed flatbed reacto reactorr with with several several front front and and back back chambers chambers sig- significantlynificantly limited limited an an effective effective mixing mixing and and gas distribution.distribution. However,However, itsits performance performance couldcould be be substantially substantially improved improved by by operation operation in in a a circulation circulation mode mode with with a a pump. pump. Under Under lessless favorable favorable cloudy cloudy conditions conditions (Table (Table3), 3), the the CPC CPC and and flatbed flatbed reactors reactors achieved achieved complete complete conversionsconversions afterafter 3030 h h of of exposure exposure over over 4 4 days. days. In In contrast, contrast, both both the the horizonal horizonal and and the the verticalvertical tubetube reactorsreactors gave incomplete incomplete conver conversionssions of of 66% 66% at atthe the end end of the of thecampaign campaign after after33 h. 33 The h. Thesophisticated sophisticated design design of the of CPC the CPCconcentrator concentrator and the and large, the large, fully fullyexposed exposed aper- apertureture of the of the flatbed flatbed were were considered considered benefici beneficialal and and gave gave superior conversionsconversions basedbased on on time and area. The small surface areas of the tube reactors naturally prevented an efficient harvesting of diffuse sunlight. The PROPHIS loop was not used under cloudy conditions due to its dependence on direct solar radiation. An economic evaluation study for the manufacturing of rose oxide (15) on a 100 t/year scale in Germany has been described by Monnerie and Ortner [95]. For this, initial investment and subsequent operation costs for solar vs. lamp-driven photooxygenation processes were compared and assessed. The study suggested that the solar operation was significantly more profitable. Molecules 2021, 26, x FOR PEER REVIEW 12 of 28

time and area. The small surface areas of the tube reactors naturally prevented an efficient harvesting of diffuse sunlight. The PROPHIS loop was not used under cloudy conditions due to its dependence on direct solar radiation.

Table 3. Reactor and experimental details of the solar campaign.

reactions 9 reactions 10 reactions 11 reactions 12 reactions 13 reactor PROPHIS CPC flatbed horizontal tubes vertical tubes CF 32 suns 1 sun 1 sun 1 sun 1 sun aperture 32 m2 3 m2 1.5 m2 not given not given volume 80 L (37 L 1) 40 L (26 L 1) 21 L 1 25 L (11 L 1) 20 L (10 L 1) operation circulation circulation static circulation circulation mixing static mixer 7 × 180° turns air bubbling 9 × 180° turns air bubbling citronellol (12) 8 L (43.9 mol) 4 L (22 mol) 2.1 L (11.5 mol) 2.5 L (13.7 mol) 2 L (11 mol) isopropanol 72 L 36 L 18.9 L 22.5 L 18 L rose bengal 2 40 g 20 g 10.5 g 12.5 g 10 g O2 source oxygen air air air air time (sunny/cloudy) 2½ hours 3/not op- 7 h/30 h 15 h/30 h 12 h/33 h 13 h/33 h 4 erated >95% 3/not oper- conversion of 12 5 >95%/>95% >95%/>95% >95%/>65% >95%/>65% ated 1 Exposed volumes (all tubes). 2 Added in several dosages. 3 Reaction 6 from Table 2. 4 Illumination time under sunny/cloudy conditions. 5 Determined by GC (vs. internal standard) after subsequent reduction with Na2SO3.

An economic evaluation study for the manufacturing of rose oxide (15) on a 100 t/year scale in Germany has been described by Monnerie and Ortner [95]. For this, initial investment and subsequent operation costs for solar vs. lamp-driven photooxygenation Molecules 2021, 26, 1685 processes were compared and assessed. The study suggested that the solar operation12 of 27 was significantly more profitable. Dincalp and Icli investigated diimide-sensitized photooxidations of (-)-citronellol (12) in concentratedDincalp and Icli sunlight investigated at 90–100 diimide-sensitized suns (Scheme photooxidations8 and Table 4) [96]. of (-)-citronellol Solar exposures were(12 )performed in concentrated with sunlightacetonitrile at 90–100 solutions suns (Schemeof 12 (0.28 andM) that Table were4)[ 96 purged]. Solar with exposures air during experimentalwere performed runs. with Reaction acetonitrile mixtures solutions were of pl12aced(0.2 in M) a thatphotoreactor were purged vessel with and air during positioned in theexperimental focal point runs. of Reactiona dish concentrator mixtures were (Figure placedin 3). a photoreactorSamples were vessel withdrawn and positioned during or afterin theillumination focal point ofand, a dish after concentrator further thermal (Figure3 ).reduction Samples were with withdrawn Na2SO3, duringwere analyzed or after by GC–MS.illumination and, after further thermal reduction with Na2SO3, were analyzed by GC–MS.

SchemeScheme 8. Diimide-sensitized 8. Diimide-sensitized photooxidations photooxidations of of (-)- 12 andand subsequent subsequent conversions conversions to rose to rose oxide oxide (15). (15).

TableUsing 4. Experimental ABIPER (8a details) as a of photosensitizer diimide-sensitized (reaction photooxidations 14), the of (-)- conversion12. of (-)-citronellol (12) increased with prolongedReaction 14illumina Reactiontion and 15 after 2 Reactionh, rose oxide 16 15 Reactionwas obtained 17 as a diastereoisomericsensitizer mixture ABIPER in (a8a total) amount BUNAP of (16 ca.) 82%, ABIPER next to ( 8athe) regioisomeric ABIPER (8a )diols 14a and b andadditive residual 12 with none approx. 6% none and 3%, respectively. CuPiv2 When FeMyrthe naphthalene3 exposure time 1 2 h 1 h 2 h 2 h residual 12 2 3% 1% 81% 36% 14a/14b 2 6% 20% 2% 5% 15 (cis/trans 3) 2 76%/6% 54%/23% 11%/5% 53%/6% 1 2 3 Maximum exposure times. Determined by GC after subsequent reduction with Na2SO3. Diasteroisomers.

Using ABIPER (8a) as a photosensitizer (reaction 14), the conversion of (-)-citronellol (12) increased with prolonged illumination and after 2 h, rose oxide 15 was obtained as a diastereoisomeric mixture in a total amount of ca. 82%, next to the regioisomeric diols 14a and b and residual 12 with approx. 6% and 3%, respectively. When the naphthalene diimide sensitizer BUNAP 16 was employed (reaction 15), conversions of 12 again increased with extended exposure times. The photoproduct obtained after 1 h of exposure to concentrated sunlight contained ca. 77% of rose oxide (15), 20% of diols 14a and b and 1% of unreacted (-)-citronellol (12). Despite the high sunlight concentration factor, both photosensitizers remained visibly photostable throughout the experiments. In line with earlier investigations, an electron transfer mechanism involving the super oxide anion •– 1 radical (O2 ) was postulated instead of a singlet oxygen ( O2) pathway. Addition of cupric pivalate (CuPiv2; reaction 16) or ferric myristate (FeMyr3; (C13H27CO2)3Fe; reaction 17) retarded the reaction and solar illuminations for 2 h reached conversions of 12 of just approximately 19% and 64%, respectively. The diasteroisomeric ratios for cis- and trans-15 varied depending on experiment, but no reasonable explanation was given to rationalize these ﬁndings.

4.3. Photooxygenations of β-Pinene for the Production of Myrtenol The Schenck ene-photooxygenation of β-pinene (17), followed by successive reduction of the hydroperoxide intermediate 18, yields myrtenol (19), with myrtenal (20) as a common by-product (Scheme9)[ 97]. Myrtenol is used as a fragrance ingredient in a variety of cosmetic, household and detergent products with an annual demand of <0.1 tons reported in 2008 [98]. Using a modiﬁed photooxygenation procedure, myrtenal (20) can be obtained as the main product instead [99]. Molecules 2021, 26, x FOR PEER REVIEW 13 of 28

diimide sensitizer BUNAP 16 was employed (reaction 15), conversions of 12 again increased with extended exposure times. The photoproduct obtained after 1 h of exposure to concentrated sunlight contained ca. 77% of rose oxide (15), 20% of diols 14a and b and 1% of unreacted (-)-citronellol (12). Despite the high sunlight concentration factor, both photosensitizers remained visibly photostable throughout the experiments. In line with earlier investigations, an electron transfer mechanism involving the super oxide anion radical (O2•–) was postulated instead of a singlet oxygen (1O2) pathway. Addition of cupric pivalate (CuPiv2; reaction 16) or ferric myristate (FeMyr3; (C13H27CO2)3Fe; reaction 17) retarded the reaction and solar illuminations for 2 h reached conversions of 12 of just approximately 19% and 64%, respectively. The diasteroisomeric ratios for cis- and trans-15 varied depending on experiment, but no reasonable explanation was given to rationalize these findings.

Table 4. Experimental details of diimide-sensitized photooxidations of (-)-12.

reaction 14 reaction 15 reaction 16 reaction 17 sensitizer ABIPER (8a) BUNAP (16) ABIPER (8a) ABIPER (8a) additive none none CuPiv2 FeMyr3 exposure time 1 2 h 1 h 2 h 2 h residual 12 2 3% 1% 81% 36% 14a/14b 2 6% 20% 2% 5% 15 (cis/trans 3) 2 76%/6% 54%/23% 11%/5% 53%/6% 1 Maximum exposure times. 2 Determined by GC after subsequent reduction with Na2SO3. 3 Diasteroisomers.

4.3. Photooxygenations of β-Pinene for the Production of Myrtenol The Schenck ene-photooxygenation of β-pinene (17), followed by successive reduction of the hydroperoxide intermediate 18, yields myrtenol (19), with myrtenal (20) as a common by-product (Scheme 9) [97]. Myrtenol is used as a fragrance ingredient in a variety of cosmetic, household and detergent products with an annual demand of <0.1 tons Molecules 2021, 26, 1685 13 of 27 reported in 2008 [98]. Using a modified photooxygenation procedure, myrtenal (20) can be obtained as the main product instead [99].

β SchemeScheme 9. Photooxygenation 9. Photooxygenation of of β-pinene-pinene ( (1717)) and subsequentsubsequent conversion conversion to myrtenolto myrtenol (19). (19). Several solar photooxygenations of 17 have been independently reported and selected detailsSeveral are compiledsolar phot inooxygenations Table5. The rose of bengal-sensitized 17 have been independently reaction of a 5 vol% reported solution and se- lectedof β details-pinene are (17 )compiled in isopropanol in Table has 5. been The conductedrose bengal-sensitized at the German reaction Aerospace of a Center 5 vol% so- lutionDLR of in β Germany-pinene ( in17 a) in CPC isopropanol reactor (Figure has6 beena). The conducted reaction proceeded at the German very slowly Aerospace and a Cen- ter conversionDLR in Germany of 17 of in approximately a CPC reactor 95% (Figure was achieved 6a). The only reaction after exhaustiveproceeded exposure very slowly for and a conversion9 days (reaction of 17 18). of approximately 95% was achieved only after exhaustive exposure for 9 days (reactionTo reach higher 18). space-time yields, the reaction was transferred to the solar furnace at 2 theTo DLR reach (Figure higher7a) [ 100space-time]. In this facility,yields, sunlight the reaction is reflected was bytransferred a single sun-tracking to the solar 52 mfurnace heliostat onto a concentrator made from 147 spherical mirrors (aperture of 39.1 m2) that at the DLR (Figure 7a) [100]. In this facility, sunlight is reflected by a single sun-tracking directs the resulting sun beam on the experimental setup inside the laboratory room [61]. 2 2 52 Solarm heliostat photooxygenations onto a concentrator of 17 were made performed from in147 a specializedspherical mirrors circulating (aperture reactor of fitted 39.1 m ) thatwith directs an IR-filter the resulting cell and sun a reaction beam on cell, the the ex laterperimental with a totalsetup volume inside ofthe approx. laboratory 5.1 L. room [61].An Solar initial photooxygenations experiment was conducted of 17 were in non-flammable performed in chloroform a specialized under circulating mainly sunny reactor fittedconditions with an using IR-filter only cell 67 of and the a spherical reaction mirrors cell, the (aperture later with of 18 a mtotal2) and volume tetraphenylpor- of approx. 5.1 L. phyrineAn initial (TPP) experiment as a sensitizer. was Due conducted to sensitizer in bleaching non-flammable and technical chloroform difficulties under with themainly oxygen supply, the reaction was stopped after less than 3 h. After reduction with thiourea, a conversion of 17 of approx. 30% was determined by GC analysis. The solar reactor was consequently modified to allow for the usage of flammable solvents and a better oxygen feeding (Figure7b). The improved device was then applied to rose bengal-sensitized photooxygenations of β-pinene (17) in ethanol. To account for possible quenching effects by breakdown products of the sensitizer rose bengal during exhaustive illumination [101], initial experiments were stopped after partial conversions of 17 of approx. 45% and 65%, respectively. After reduction with Na2SO3 and aqueous workup to remove any degradation products, the crude mixtures containing unreacted 17 and myrtenol (19) were subjected to further photooxygenation. Following this approach, conversions of 97% (by GC) were achieved after 14 h of exposures under largely sunny conditions. The rose bengal-sensitized photooxygenation of 17 (0.52 M) under comparable conditions without interruption and in between workup also achieved near complete conversion of 97% after 14 h of illumination (reaction 19). Quenching or retarding effects by potential photodegradation products were Molecules 2021, 26, x FOR PEER thusREVIEW ruled out. From this experimental run, a high space-time yield of 3.6×10-2 mol/(L15· h)of 28

was furthermore determined. However, the high construction, operation and maintenance costs of solar furnaces prevent their widespread application for preparative photochemistry.

(a) (b)

FigureFigure 7. (a) 7. Solar (a) Solar furnace furnace at the at German the German Aerospace Aerospace Center Center (DLR) (DLR in Cologne-Porz,) in Cologne-Porz, Germany, Germany, around around 2005 (taken 2005 from (taken [39 from]). (b) Custom-built[39]). (b) Custom-built photooxygenation photooxygenation reactor, around reacto 2001r, around (taken 2001 from (taken [39]). from [39]).

4.4. Photooxygenations of α-Thujene for the Production of Trans-Sabinene Hydrate The important and valuable flavor compound trans-sabinene hydrate (24; also known as trans-4-thujanol) is synthesized industrially from α- or 3-thujene (21) (Scheme 10) [55,104,105]. Initial photooxygenation of 21 and subsequent reduction of the regioisomeric hydroperoxides 22a and b yields (-)-trans-4-hydroxy-β-thujene (23a) and (-)-cis-sabinol (23b) in a ratio of approx. 3:1. Catalytic hydrogenation of 23a produces the desired trans- sabinene hydrate (24), while alternative oxidation furnishes the naturally occurring and somewhat toxic (-)-umbellulone (25).

Scheme 10. Photooxygenation of α-thujene (21) and subsequent conversions to trans-sabinene hydrate (24) or (-)-umbellulone (25).

Molecules 2021, 26, 1685 14 of 27

Table 5. Reactor and experimental details of selected solar photooxygenations of β-pinene (17).

Reaction 18 Reaction 19 [100] Reaction 20 [102] reactor CPC DLR solar furnace SOLARIS CF 1 sun up to 4,800 suns 20 suns (42 suns 1) volume 40 L (26 L 2) 5.1 L (2.03 L 2) 70 L (30 L 2) β-pinene (17) 2 L (12.8 mol) 0.29 L (1.84 mol) 6.2 L (37.7 mol) Molecules 2021, 26, x FOR PEER REVIEW solvent i 3.5 L (EtOH) i 15 of 28 38 L ( PrOH) 35 L ( PrOH) rose bengal not reported 40 g 3 21.77 g 3 exposure time 9 days 14 h 3 days (ca. 27 h) conversion of 17 4 95% 97% ca. 56% 1 Theoretical concentration factor without optical losses. 2 Exposed volumes of reactor. 3 Added in several 4 dosages. Determined by GC directly or after subsequent reduction with Na2SO3.

Esser furthermore investigated the solar transformation of β-pinene (17) with moderately concentrated sunlight at the Plataforma Solar de Almería (PSA) in Spain in 1992 [102]. The experiment was conducted in the former SOLARIS (SOLAR photochemical synthesIS of fine chemicals) facility, an earlier version of the PROPHIS loop (Figure1c) [ 103]. This parabolic trough reactor used aluminized polymeric film reflectors, had an effective concentration factor of 20 suns in practice (theoretical 42 suns) and an operating capacity of 35-70 L. Utilizing one trough (aperture of 7.29 m2), a solution of 17 and rose bengal (in 10 portions) in 35 L of isopropanol was passed through the loop at approx. 42 L/min while oxygen was injected into the reagent stream. The reaction temperature was maintained at roughly 20 ◦C and the progress of the reaction was monitored by GC analysis (reaction 20). After (a) approximately 27 h of exposure over three consecutive days with( goodb) to excellent sunlight conditions, a conversion of β-pinene (17) of approx. 56% was reached, corresponding to an Figure 7. (a) Solar furnace at the German Aerospace Center (DLR) in Cologne-Porz, Germany, around 2005 (taken from average conversion rate of 2.2%/h (or 0.84 mol/h). Subsequent solar thermal reduction [39]). (b) Custom-built photooxygenation reactor, around 2001 (taken from [39]). with aqueous sodium sulfite (Na2SO3) in the SOLARIS loop furnished a crude product mixture containing ca. 40% of residual 17, 44% of myrtenol (19), 7% of myrtenal (20) and 4.4. Photooxygenations1% of hydroperoxide of18 α(by-Thujene GC), respectively. for the Production of Trans-Sabinene Hydrate

The4.4. Photooxygenations important and of valuableα-Thujene flavor for the Productioncompound of Trans-Sabinene trans-sabinene Hydrate hydrate (24; also known as trans-4-thujanol)The important is and synthesized valuable ﬂavor industrially compound trans from-sabinene α- or hydrate 3-thujene (24; also (21 known) (Scheme 10) [55,104,105].as trans Initial-4-thujanol) photooxygenation is synthesized of industrially21 and subsequent from α reduction- or 3-thujene of the regioisomeric (21) hydroperoxides(Scheme 10)[55 22a,104 and,105] .b Initial yields photooxygenation (-)-trans-4-hydroxy- of 21 andβ-thujene subsequent (23a reduction) and (-)- ofcis-sabinol (23b)the in regioisomerica ratio of approx. hydroperoxides 3:1. Catalytic22a and hydrogenationb yields (-)-trans-4-hydroxy- of 23a producesβ-thujene (the23a) desired and trans- (-)-cis-sabinol (23b) in a ratio of approx. 3:1. Catalytic hydrogenation of 23a produces the sabinenedesired hydratetrans-sabinene (24), while hydrate alternative (24), while oxidation alternative oxidationfurnishes furnishes the naturally the naturally occurring and somewhatoccurring toxic and (-)-umbellulone somewhat toxic (-)-umbellulone (25). (25).

α Scheme 10.Scheme Photooxygenation 10. Photooxygenation of α-thujene of -thujene (21) and (21) subsequent and subsequent conversions conversions to to transtrans-sabinene-sabinene hydrate hydrate (24 ()24 or) or (-)- (-)-umbel- umbellulone (25). lulone (25).

Molecules 2021, 26, 1685 15 of 27

In 1992, Esser conducted a series of seven solar photooxygenations of α-thujene (21) in isopropanol in the SOLARIS reactor equipped with one trough (Figure8 and Table6) [102,106]. Technical grade 21, which contained α-pinene as impurity, with a purity of ca. 87% was employed. α-Pinene also underwent photooxygenations and its corresponding reduced photoproducts were thus detected and quantified [97]. Methylene 1 blue (MB) or rose bengal (RB) were investigated as O2 photosensitizers. Both showed significant decomposition in concentrated sunlight and hence required regular refeeding during the experiments. Depending on conversion rates and weather conditions, fluid and oxygen gas flows were adjusted between 35–45 L/min and 1.0–5.5 L/min, respectively. For reactions 21–26, the reaction mixtures were cooled to approximately 20 ◦C during circulation within the reactor loop. Only for reaction 27, the temperature was increased to Molecules 2021, 26, x FOR PEER REVIEW ◦ 16 of 28 approx. 30 C instead. Based on the total solar radiation recorded, the corresponding moles of photons received in the absorption range of the respective sensitizer, and hence solar photon yields (ηs) were calculated. All solar photooxygenations proceeded efficiently and 1 thermalfurnished reductions high to nearwere complete directly conversions performed of inside21 of >87%the SOLARIS after exposure loop timesat temperatures of 4 2 -11 h. >50 °C.Due Except to the for instability reaction of 22, the hydroperoxidehigh selectivities intermediates for (-)-trans22a-4-hydroxy-and b, their subsequentβ-thujene solar(23a) for- thermal reductions were directly performed inside the SOLARIS loop at temperatures mation of ≥72% were achieved. The best overall results with respect to exposure time, >50 ◦C. Except for reaction 22, high selectivities for (-)-trans-4-hydroxy-β-thujene (23a) conversion,formation selectivity of ≥72% were and achieved. solar photon The best yield overall were results accomplished with respect with to exposure rose bengal time, at 20 °C conversion,(reactions selectivity25 and 26). and From solar the photon combined yield were crude accomplished product withmixture rose bengalof all experiments, at 20 ◦C the(reactions desired (-)- 25 andtrans 26).-4-hydroxy- From the combinedβ-thujene crude(23a) productwas isolated mixture in of 78% all experiments,purity (by GC) the and 65%desired yield, (-)- bothtrans exceeding-4-hydroxy- theβ-thujene reported (23a valu) wases isolated for lamp-driven in 78% purity processes. (by GC) andThis 65% was ex- plainedyield, by both the exceeding smaller theUV-content reported valuesof the forreflected lamp-driven natural processes. sunlight. This was explained by the smaller UV-content of the reflected natural sunlight.

(a) (b) FigureFigure 8. Former 8. Former SOLARIS SOLARIS loop loop at atthe the Plataforma Plataforma Solar Solar de de Almería Almería (PSA)(PSA) near near Tabernas, Tabernas, Spain, Spain, in 1992. in 1992. (a) Used (a) Used for a for a methylenemethylene blue blueand and(b) rose (b) rose bengal-sensitized bengal-sensitized photooxygenation. photooxygenation.

4.5.Table Photooxygenations 6. Experimental of Furfural details of and photooxidations Furfuraldiethylacetal of 21. to 5-Hydroxy- and 5-Alkoxyfuranones reaction 21 reactionThe 22 photooxygenation reaction 23 of reacti furfuralon (2426) in reaction various alcohols 25 reaction was ﬁrst 26 described reaction by 27 α-thujene 2.7 kg Schenck,5.42 kg who isolated5.45 thekg corresponding~3.3 kg 5-alkoxyfuranone 5.63 kg derivatives5.6428 kg(Scheme 115.63)[106 kg, (21) 1 (17.2 mol) (34.5107]. Itmol) was later(34.6 shown mol) that 5-hydroxyfuranone(20.9 mol) ((35.827) is mol) formed as(35.9 the initial mol) photoproduct, (35.8 mol) iPrOH 44 L which 38 L can be subsequently 35 L isolated ~32 in L yields of ≤90% 35 L [ 108,109]. The 35 pseudo-acidL 27 35 isL rapidly converted to 28 at elevated temperatures >70 ◦C, which were commonly exceeded 5.26 g 9.61 g 11.11 g 12.29 g 12.82 g 9.02 g 6.95 g sensitizer 2 in earlier studies. Hydroxyfuranone (27) has pesticidal properties but also serves as a (MB) (MB) (MB) (MB) (RB) (RB) (RB) versatile C4 building block in organic synthesis [110]. Longer-chain 5-alkoxyfuranones, ~7 h prepared~11 h by acid-catalyzed7 h (trans)esteriﬁcation,~7 h have6 beenh described5½ ashours fragrances 4½ [111 hours]. time 3 (1 day) (2 days) (1 day) (1 day) (1 day) (1 day) (1 day) total radiation 16.5 kWh 37.9 kWh 40.8 kWh 43.0 kWh 29.9 kWh 27.5 kWh 24.2 kWh 4 (110 mol) (271 mol) (287 mol) (306 mol) (123 mol) (113 mol) (106 mol) ηs 5 14% 12% 12% 7% 29% 31% 33% conversion 6 88% 97% 96% 97% ~100% 99% 98% 21 7 12% 2% 3% 2% <1% 1% 1% 23a 7 61% 56% 63% 66% 65% 67% 61% 23b 7 7% 11% 11% 10% 10% 11% 13% selectivity 8 82% 67% 75% 78% 75% 78% 72% 1 Technical grade (86.5%). 2 Sensitizer added in several dosages. 3 Illumination time (over number of days). 4 Calculated molar radiation up to 700 nm (MB) or 600 nm (RB). 5 Solar photon yield. 6 Conversion of 21 by GC. 7 Product composition by GC after reduction with Na2SO3 (by-products excluded). 8 Selectivity for 23a.

4.5. Photooxygenations of Furfural and Furfuraldiethylacetal to 5-Hydroxy- and 5- Alkoxyfuranones The photooxygenation of furfural (26) in various alcohols was first described by Schenck, who isolated the corresponding 5-alkoxyfuranone derivatives 28 (Scheme 11) [106,107]. It was later shown that 5-hydroxyfuranone (27) is formed as the initial photoproduct, which can be subsequently isolated in yields of ≤90% [108,109]. The pseudo-acid 27 is rapidly converted to 28 at elevated temperatures >70 °C, which were commonly exceeded in earlier studies. Hydroxyfuranone (27) has pesticidal properties but also serves

Molecules 2021, 26, 1685 16 of 27

Table 6. Experimental details of photooxidations of 21.

Reaction 21 Reaction 22 Reaction 23 Reaction 24 Reaction 25 Reaction 26 Reaction 27 α-thujene 2.7 kg 5.42 kg 5.45 kg ~3.3 kg 5.63 kg 5.64 kg 5.63 kg (21) 1 (17.2 mol) (34.5 mol) (34.6 mol) (20.9 mol) (35.8 mol) (35.9 mol) (35.8 mol) iPrOH 44 L 38 L 35 L ~32 L 35 L 35 L 35 L 5.26 g 9.61 g 11.11 g 12.29 g 12.82 g 9.02 g 6.95 g sensitizer 2 (MB) (MB) (MB) (MB) (RB) (RB) (RB) ~7 h ~11 h 7 h ~7 h 6 h 5 1 h 4 1 h time 3 2 2 (1 day) (2 days) (1 day) (1 day) (1 day) (1 day) (1 day) 16.5 kWh 37.9 kWh 40.8 kWh 43.0 kWh 29.9 kWh 27.5 kWh 24.2 kWh total radiation 4 (110 mol) (271 mol) (287 mol) (306 mol) (123 mol) (113 mol) (106 mol) 5 Molecules 2021, 26, x ηFORs PEER REVIEW14% 12% 12% 7% 29% 31% 33% 17 of 28 conversion 6 88% 97% 96% 97% ~100% 99% 98% 21 7 12% 2% 3% 2% <1% 1% 1% 23a 7 61% 56% 63% 66% 65% 67% 61% 23b 7 7% 11% 11% 10% 10% 11% 13% selectivity 8 82% 67% 75% 78% 75% 78% 72% as a versatile C4 building block in organic synthesis [110]. Longer-chain 5-alkoxy- 1 Technical grade (86.5%). 2 Sensitizer added in several dosages. 3 Illumination time (over number of days). 4 Calculated molar radiation up to 700 nm (MB) or 600furanones, nm (RB). 5 Solar prepared photon yield. by6 Conversionacid-catalyzed of 21 by GC. (trans)esterification,7 Product composition by GC have after reduction been withdescribed as fra- 8 Na2SO3 (by-products excluded).grancesSelectivity [111]. for 23a.

Scheme 11.Scheme Photooxygenation 11. Photooxygenation of furfural of furfural (26 (26) to) to 5-hydroxyfuranone 5-hydroxyfuranone (27) ( and27) subsequentand subsequent conversion conversion to 5-alkoxyfuranones to 5-alkoxyfuranones (28). (28). The solar exposure of a solution of 26 (100 g) and eosin (1 g) in ethanol (1 L) in a large 1 conical flask has been described in a patent in 1953 [108]. After 1 2 days in direct sunlight, The5-ethoxyfuranone solar exposure (28; Rof = a Et) solution was isolated of 26 in (100 a yield g) ofand 50% eosin (66.7 g).(1 g) in ethanol (1 L) in a large conical flaskIn 1991 has and been 1992, described sixteen large-scale in a patent photooxygenations in 1953 [108]. After were subsequently1½ days in direct inves- sunlight, 5-ethoxyfuranonetigated in the SOLARIS (28; R = loop Et) was equipped isolated with in a single a yield trough of 50% (Figure (66.78)[ g).47, 102]. Process conditions were systematically varied and included alterations in the starting material In(26 1991vs. 29 and), photosensitizer 1992, sixteen (RB large-scale vs. MB), temperature, photooxygenations fluid flow rates were and possiblesubsequently addi- investi- gatedtives. in the Five SOLARIS selected examplesloop equipped from this with solar a campaignsingle trough are summarized (Figure 8) in [47,102]. Table7[ 47Process]. con- ditionsThe were oxygen systematically gas flow rates varied were adjusted and included based on alterations the weather conditionsin the starting and reaction material (26 vs. 29), photosensitizerprogress and ranged (RB between vs. MB), 0.7 temperature, and 5.5 L/min. fluid Methylene flow rates blue wasand identifiedpossible asadditives. the Five preferred photosensitizer due to its superior stability under the weakly acidic reaction selectedconditions. examples Despite from varying this weathersolar campaign conditions, allare solar summarized reactions furnished in Table near 7 complete[47]. The oxygen gas flow rates were adjusted 1based on the weather conditions and reaction progress and conversions of >95% after 3 4 -16 h of illumination. The reaction was found to be very rangedrobust between and partially 0.7 and converted 5.5 L/min. reaction Methylene mixtures couldblue bewas kept identified in the reactor as the loop preferred during photo- sensitizerbad weather due to periods.its superior The addition stability of under concentrated the weakly hydrochloric acidic acid reaction (reaction conditions. 31) was Despite varyingfound weather beneficial conditions, due to the rapidall solar formation reactions of furfuraldiethylacetal furnished near ( 29complete), which reacted conversions of more rapidly with singlet oxygen than furfural (26) itself. This was demonstrated when >95%diethylacetal after 3¼-1629 h wasof illumination. used as the starting The reaction material (reaction was found 32). Whileto be very the selectivity robust and for partially convertedthe formation reaction of mixtures hydroxyfuranone could be (27 kept) was in high the duringreactor the loop actual during photooxygenation, bad weather periods. The additionsignificant of conversion concentrated into ethoxyfuranone hydrochloric (28 ac;id R = (reaction Et) occurred 31) during was found further workupbeneficial due to the rapidandisolation. formation No of attempts furfuraldiethylacetal to isolate the more (29 desirable), which27 reactedfrom these more mixtures rapidly were with singlet consequently made. However, the acid-catalyzed (trans)esterification of a crude mixture oxygen(ca. than 250 g) furfural containing (2627)and itself.28 (R This = Et) was with demonstratedn-hexanol to hexyloxyfuranone when diethylacetal (28; R = Hex) 29 was used as thein starting 79% yield material has been reported(reaction in a32). patent While [111 ].the selectivity for the formation of hydroxyfuranone (27) was high during the actual photooxygenation, significant conversion into ethoxyfuranone (28; R = Et) occurred during further workup and isolation. No attempts to isolate the more desirable 27 from these mixtures were consequently made. However, the acid-catalyzed (trans)esterification of a crude mixture (ca. 250 g) containing 27 and 28 (R = Et) with n-hexanol to hexyloxyfuranone (28; R = Hex) in 79% yield has been reported in a patent [111].

Table 7. Experimental details of selected photooxidations of furane derivatives 26 and 29.

reaction 28 reaction 29 reaction 30 reaction 31 reaction 32 2.06 kg 2.14 kg 4.32 kg 3.61 kg kg 3.61 kg furane 21.5 mol (26) 22.3 mol (26) 45.0 mol (26) 37.6 mol (26) 21.2 mol (29) ethanol 35 L 35 30 35 35 sensitizer 1 6.1 g (MB) 11.2 g (RB) 5.1 g (MB) 9.0 g (MB) 6.0 g (MB) additive none none none c. HCl (5 mL) none temperature 8 °C 20 °C 20 °C 20 °C 20 °C fluid flow 35 L/min 55 L/min 55 L/min 45 L/min 45 L/min 16 h ~11 h ~12½ hours 5½ hours ~3¼ hours time 2 (3 days) (2 days) (2 days) (1 day) (1 day) 22.2 kWh 54.3 kWh 61.9 kWh 32.4 kWh 16.9 kWh total radiation 3 140 mol 413 mol 436 mol 226 mol 114 mol ηs 4 15% 5% 10% 16% 18% conversion 5 98% 99% ~100% 98% 95% 27:28 6 42:1 133:1 58:1 98:1 390:1

Molecules 2021, 26, 1685 17 of 27

Table 7. Experimental details of selected photooxidations of furane derivatives 26 and 29.

Reaction 28 Reaction 29 Reaction 30 Reaction 31 Reaction 32 2.06 kg 2.14 kg 4.32 kg 3.61 kg 3.61 kg furane 21.5 mol (26) 22.3 mol (26) 45.0 mol (26) 37.6 mol (26) 21.2 mol (29) ethanol 35 L 35 30 35 35 sensitizer 1 6.1 g (MB) 11.2 g (RB) 5.1 g (MB) 9.0 g (MB) 6.0 g (MB) additive none none none c. HCl (5 mL) none temperature 8 ◦C 20 ◦C 20 ◦C 20 ◦C 20 ◦C ﬂuid ﬂow 35 L/min 55 L/min 55 L/min 45 L/min 45 L/min 16 h ~11 h ~12 1 h 5 1 h ~3 1 h time 2 2 2 4 (3 days) (2 days) (2 days) (1 day) (1 day) total 22.2 kWh 54.3 kWh 61.9 kWh 32.4 kWh 16.9 kWh 3 Molecules 2021, 26, x FOR PEER REVIEWradiation 140 mol 413 mol 436 mol 226 mol 114 mol18 of 28 4 ηs 15% 5% 10% 16% 18% conversion 5 98% 99% ~100% 98% 95% 27:28 6 42:1 133:1 58:1 98:1 390:1 1 Sensitizer added in several1 Sensitizerdosages. added 2 Illumination in several dosages.time (over2 Illumination number of time days). (over 3number Calculated of days). molar3 Calculated radiation molar up to radiation 700 up nm. 4 Solar photon yield. 5 Conversionto 700 nm. 4 Solarof 26 photonor 29 by yield. GC.5 6Conversion Ratio of 27 of vs.26 or28 29(Rby = Et) GC. by6 Ratio GC. of 27 vs. 28 (R = Et) by GC.

4.6. Photooxygenations of 1,5-Dihydroxynaphthalene to Juglone Juglone oror 5-hydroxy-1,4-naphthoquinone 5-hydroxy-1,4-naphthoquinone (31 (31) is) ais naturally a naturally occurring occurring compound compound that isthat used is used as a herbicide,as a herbicide, dye, coloringdye, coloring agent agent and versatile and versatile building building block chemicalblock chemical [112]. It[112]. can beIt can obtained be obtained in good in yields good fromyields 1,5-dihydroxynaphthalene from 1,5-dihydroxynaphthalene (30) through (30) through photooxygenation photooxy- (Schemegenation 12(Scheme)[113]. 12) [113].

Scheme 12. Photooxygenation of 1,5-dihydroxynaphthalene ((3030)) toto juglonejuglone ((3131).).

Small laboratory-scale photooxygenations of 30 (0.5-2 g) havehave beenbeen conductedconducted byby Oelgemöller and co-workers co-workers in in custom-built custom-built parabolic parabolic trough trough reactors reactors (Table (Table 8).8). Initial Ini- tial experiments were performed in a horizontal trough reactor (Pyrex tube diameter of experiments were performed in a horizontal trough reactor (Pyrex tube diameter of 1.2 1.2 cm) equipped with holographic mirror elements and using isopropanol as a solvent cm) equipped with holographic mirror elements and using isopropanol as a solvent (Fig- (Figure9a) [91] . The reactor had an optimal reflectivity range for the usage of rose bengal ure 9a) [91]. The reactor had an optimal reflectivity range for the usage of rose bengal of of 480–620 nm [92,93], aperture of 0.188 m2 and a concentration factor of approx. 15 suns. 480–620 nm [92,93], aperture of 0.188 m2 and a concentration factor of approx. 15 suns. Oxygen gas was injected into the circulating reagent flow in a Y-connector, thus generating Oxygen gas was injected into the circulating reagent flow in a Y-connector, thus generat- an uneven gas-liquid flow. Conversion rates of ≥83% were achieved after illuminations ing an uneven gas-liquid flow. Conversion rates of ≥83% were achieved after illumina- for 3 h and 8 h and juglone (31) was isolated in yields of 54% and 79%, respectively tions for 3 h and 8 h and juglone (31) was isolated in yields of 54% and 79%, respectively (reactions 33 and 34). (reactions 33 and 34). Using the identical reactor equipped with a polished aluminum reflector instead, the Using the identical reactor equipped with a polished aluminum reflector instead, the same authors tested different solvents (isopropanol vs. acetone) and sensitizer materials (RB same authors tested different solvents (isopropanol vs. acetone) and sensitizer materials vs. MB; soluble vs. solid-supported) [114]. Applying a standard solution of 2 g of diol 30 in 250(RB mLvs. MB; of solvent soluble and vs. asolid-supported) fixed exposure time [114]. of A fourpplying hours, a standard the experiments solution conducted of 2 g of diol in either30 in 250 solvent mL withof solvent soluble and rose a bengalfixed exposure showed the time best ofperformances four hours, the (reactions experiments 35 and con- 36). Bothducted runs in either achieved solvent high with conversions soluble ofrose≥ 93%bengal and showed furnished the juglonebest performances (31) in isolated (reactions yields of35 75%and and36). 79%,Both respectively.runs achieved The high solid-supported conversions sensitizingof ≥93% and materials furnished gave juglone significantly (31) in lowerisolated conversion yields of rates75% ofand 46-70% 79%, afterrespective the samely. The illumination solid-supported time, presumably sensitizing due materials to the poorgave transportsignificantly and lower thus mixingconversion of these rates materials of 46-70% within after the same reactor’s illumination tube. time, presumably due to the poor transport and thus mixing of these materials within the reactor’s tube. Three additional solar photooxygenations involving diol 30 were furthermore conducted in a simple vertical parabolic trough reactor that used a Liebig condenser with a diameter of 2.4 cm as receiver tube (Figure 9b) [115]. The device used a bent polished aluminum sheet with an aperture of approx. 0.15 m2 (41 × 36 cm) and had a theoretical concentration factor of ca. 18 suns. The reaction mixture containing 30 (0.5-1.0 g) in either isopropanol or methanol (100 mL) was pumped through the outer mantle of the condenser, while cooling water was passed upwards through its central tube. After solar illuminations for effectively 2/3–4½ hours, conversion rates of 30 of ≥86% and isolated yields of juglone (31) of 46–71% were obtained. The highest conversion and isolated yield were again achieved using rose bengal and isopropanol (reaction 37).

Molecules 2021, 26, 1685 18 of 27

Table 8. Experimental details of selected photooxidations of 1,5-dihydroxynaphthalene (30).

Reaction 33 Reaction 34 Reaction 35 Reaction 36 Reaction 37 horizontal horizontal horizontal horizontal vertical reactor trough trough trough trough trough CF 15 suns 15 suns 15 suns 15 suns 18 suns reﬂector holographic holographic aluminum aluminum aluminum diol 30 2.0 g 1.0 g 2.0 g 2.0 g 0.5 g 200 mL 200 mL 250 mL 250 mL 100 mL solvent (iPrOH) (iPrOH) (iPrOH) (acetone) (iPrOH) sensitizer 0.1 g (RB) 0.1 g (RB) 0.1 g (RB) 0.1 g (RB) 0.05 g (RB) 8 h 1 3 h 1 4 h 1 4 h 1 2/ h 1 time 3 (2 days) (2 days) (1 day) (1 day) (1 day) Molecules 2021, 26, x FOR PEER REVIEW 2 19 of 28 conversion 83% >95% 93% 99% >95% yield of 31 3 54% 79% 75% 79% 71% 1 Illumination time (over number of days). 2 Conversion of 30 by GC. 3 Isolated yields of 31.

(a) (b)

FigureFigure 9. Laboratory-scale 9. Laboratory-scale parabolic parabolic trough trough concentrators concentrators at the Germanat the German Aerospace Aerospace Center (DLR)Center in (DLR) Cologne-Porz, in Cologne-Porz, Germany. Ger- (a) Horizontalmany. (a) Horizontal trough reactor trough equipped reactor withequipped holographic with holographic mirrors (the mirrors reﬂected (the greenreflecte lightd green can belight clearly can be seen), clearly 2003. seen), (b) Vertical2003. (b trough) Vertical reactor trough based reactor on a based Liebig on condenser, a Liebig condenser, 2003 (taken 2003 from (taken [115]). from [115]).

Table 8. ExperimentalThree additional details of selected solar photooxygenations photooxidations of 1,5-dihydroxynaphthalene involving diol 30 were (30 furthermore). con- reactionducted 33 in a simple reaction vertical 34 parabolic reaction trough reactor35 that reaction used a Liebig36 condenser reaction with 37 a diameter of 2.4 cm as receiver tube (Figure9b) [ 115]. The device used a bent polished reactor horizontal trough horizontal trough horizontal trough horizontal trough vertical trough aluminum sheet with an aperture of approx. 0.15 m2 (41 × 36 cm) and had a theoretical CF 15 suns 15 suns 15 suns 15 suns 18 suns concentration factor of ca. 18 suns. The reaction mixture containing 30 (0.5-1.0 g) in either reflector holographicisopropanol or methanolholographi (100c mL) wasaluminum pumped through the alum outerinum mantle of thealuminum condenser, diol 30 while2.0 g cooling water was 1.0 passedg upwards through2.0 g its central tube. 2.0 g After solar illuminations 0.5 g 200 mL 2 2001 mL 250 mL ≥ 250 mL 100 mL solvent for effectively /3–4 2 h, conversion rates of 30 of 86% and isolated yields of juglone (31) (ofiPrOH) 46–71% were obtained.(iPrOH) The highest conversion(iPrOH) and isolated(acetone) yield were again(i achievedPrOH) sensitizer 0.1using g (RB) rose bengal and 0.1 g isopropanol (RB) (reaction 0.1 g (RB) 37). 0.1 g (RB) 0.05 g (RB) 8 h 1 3 h 1 4 h 1 4 h 1 2/3 h 1 time (24.7. days) Miscellaneous Solar(2 days) Photooxygenations (1 and day) Photooxidations (1 day) (1 day) conversion 2 83%Sensitized photooxidations>95% of E-cinnamic93% acid (E-32) and 99% methyl acrylate>95% (34) in a yield of 31 3 parabolic54% dish concentrator79% (Figure3) have75% been described by79% Dincalb and Icli [ 11671%]. Aer- 1 Illuminationated solutionstime (over of numberE-32 or of34 days).in methanol 2 Conversion and of in 30 the by presence GC. 3 Isolated of sulfoamino yields of 31 perylenediimide. 8c (SULFAPER; R = -(CH2)12NHSO2OH) were exposed to concentrated sunlight (approxi- mately4.7. 21Miscellaneous suns) for up Solar to four Photoox hoursygenations and after and successive Photooxidations reduction, the product mixtures were analyzed by GC–MS coupling. Illumination of E-32 for four hours (Scheme 13) Sensitized photooxidations of E-cinnamic acid (E-32) and methyl acrylate (34) in a showed incomplete conversion and furnished a complex reaction mixture with Z-cinnamic parabolic dish concentrator (Figure 3) have been described by Dincalb and Icli [116]. Aer- acid (Z-32) and the anhydride 33 as the main products in amounts of ca. 18% and 12%, ated solutions of E-32 or 34 in methanol and in the presence of sulfoamino perylene- respectively. Methyl acrylate (34), carbonic acid (35), 2-phenylethanol (36) and benzalde- diimide 8c (SULFAPER; R = -(CH2)12NHSO2OH) were exposed to concentrated sunlight hyde (37) were identiﬁed as minor products. Additional exposure in the absence of 8c and (approximately 21 suns) for up to four hours and after successive reduction, the product under deaerated or aerated conditions delivered somewhat different product mixtures mixtures were analyzed by GC–MS coupling. Illumination of E-32 for four hours (Scheme and 2-hydroxyacetaldehyde (38) was furthermore detected. In all case examined, direct 13) showed incomplete conversion and furnished a complex reaction mixture with Z-cinnamic acid (Z-32) and the anhydride 33 as the main products in amounts of ca. 18% and 12%, respectively. Methyl acrylate (34), carbonic acid (35), 2-phenylethanol (36) and ben- zaldehyde (37) were identified as minor products. Additional exposure in the absence of 8c and under deaerated or aerated conditions delivered somewhat different product mixtures and 2-hydroxyacetaldehyde (38) was furthermore detected. In all case examined, direct photoisomerization to the isomeric Z-cinnamic acid (Z-32) was found as the dominate reaction pathway. The formation of 33, 36 and 37 was explained via photoinduced electron transfer and the involvement of the super oxide anion radical (O2•–). The small scale and the rather complex outcome of this transformation limit its usefulness in organic synthesis.

Molecules 2021, 26, x FOR PEER REVIEW 20 of 28

Molecules 2021, 26, 1685 19 of 27

photoisomerization to the isomeric Z-cinnamic acid (Z-32) was found as the dominate reaction pathway. The formation of 33, 36 and 37 was explained via photoinduced electron Molecules 2021, 26, x FOR PEER REVIEW •– 20 of 28 transfer and the involvement of the super oxide anion radical (O2 ). The small scale and the rather complex outcome of this transformation limit its usefulness in organic synthesis.

Scheme 13. SULFAPER-sensitized photooxidations of E-cinnamic acid (E-32).

The SULFAPER-sensitized reaction of methyl acrylate (34) in basic methanol was much more selective (Scheme 14) [116]. Exposure to concentrated sunlight under aerated conditions forScheme oneScheme hour13. SULFAPER-sensitized 13. furnishedSULFAPER-sensitized complete photooxidations photooxidations conversion of ofE-cinnamic ofE-cinnamic 34 and acid methylacid (E-32 ).(E-32 2-oxopropanoate). (39) was obtained as theThe only SULFAPER-sensitized product. Shorte reactionr illumination of methyl acrylate for ¼ (34hour) in basicunder methanol the same was conditions also producedmuchThe moreSULFAPER-sensitized carbonic selective (Schemeacid (35 14 ))[reaction 116in ].ca. Exposure 10%,of methyl tonext concentrated acrylate to 39 and sunlight(34) inresidual basic under methanol aerated 34 in was amounts of approximatelymuchconditions more selective69% for one and hour (Scheme 21%, furnished respec 14) complete[116].tively. Exposu conversion In rethe to of absenceconcentrated34 and methyl of the sunlight 2-oxopropanoate photosensi- under aerated (39) was obtained as the only product. Shorter illumination for 1 hour under the same tizer 8c, solar exposureconditions for for one one hour hour resulted furnished in complete a low conversion conversion ofof 34 just and4 ca. methyl 27% and2-oxopropanoate gave (39) conditionswas obtained also producedas the only carbonic product. acid (Shorte35) in ca.r illumination 10%, next to for39 and¼ hour residual under34 inthe same carbonic acid (35) and 2-hydroxyacetaldehyde (38) as the only photoproducts in amounts conditionsamounts also of approximately produced 69%carbonic and 21%, acid respectively. (35) in ca. In the10%, absence next ofto the 39 photosensitizer and residual 34 in of ca. 16% and 11%. 8cUnder, solar exposuredeareated for oneconditions hour resulted in the in a presence low conversion of SULFAPER, of just ca. 27% illumina- and gave amountscarbonic of acidapproximately (35) and 2-hydroxyacetaldehyde 69% and 21%, respec (38) astively. the only In photoproducts the absence in of amounts the photosensi- of tions for ¼ or onetizer ca. hour8c 16%, solar andyielded exposure 11%. Under methyl for deareated one 2-oxopropanoate hour conditions resulted in in the a presencelow (39 conversion) as of SULFAPER,the sole of just product illuminations ca. 27% andin gave 1 amounts of approx.carbonic for57%4 or acidand one (68%35 hour) and yieldednext 2-hydroxyacetaldehyde to methyl residual 2-oxopropanoate 34. Electron (38 (39) as)transfer as the the only sole steps productphotoproducts including in amounts in the of amounts generation of theof super ca.approx. 16% oxide and 57% anion11%. and 68% Under radical next todeareated residual (O2•–) 34were conditions. Electron postulated transfer in the steps presenceto explain including of the SULFAPER, generation formation ofillumina- •– tionsthe for super ¼ oxideor one anion hour radical yielded (O2 ) weremethyl postulated 2-oxopropanoate to explain the formation(39) as the of 39 sole. Despite product in of 39. Despite the selectivethe selective nature nature of ofthis this part particularicular transformation, transformation, the rather the rather demanding demanding technical amounts of approx. 57% and 68% next to residual 34. Electron transfer steps including the technical requirementsrequirements and the and small the small reaction reaction scale scale hinder hinder a widera wider application. application. generation of the super oxide anion radical (O2•–) were postulated to explain the formation of 39. Despite the selective nature of this particular transformation, the rather demanding technical requirements and the small reaction scale hinder a wider application.

Scheme 14. 34 Scheme 14. SULFAPER-sensitizedSULFAPER-sensitized photooxidat photooxidationsions of methyl of methyl acrylate acrylate (34 ( ). ).

The solar photooxygenation of natural rosin or collophany, which is obtained from Scheme 14. SULFAPER-sensitized photooxidations of methyl acrylate (34). The solar photooxygenationpine trees, has been of investigated natural rosin by Icli or [117 collophany,]. A solution of which rosin in methanolis obtained containing from methylene blue was exposed to concentrated sunlight in a Bomin Solar fixed focus con- pine trees, has been centratorinvestigatedThe solar capable photooxygenation by of concentrationIcli [117]. Aof factors solutionnatural of uprosin of to 2000rosin or collophany, suns. in methanol A circulating which containing reactor is obtained was from methylene blue pinewaspositioned trees,exposed has in beento the concentrated focal investigated point (1–2 cmbysunlight inIcli diameter) [117] in. A a of solutionBomin the solar ofSolar dish. rosin The fixed in reactor methanol focus consisted con- containing centrator capablemethylene ofof concentration a double-jacked blue was Pyrex exposedfactors tubing toof ofconcentratedup 30 cmto length2000 withsuns.sunlight an A outer incirculating a diameter Bomin ofSolar reactor 5.5 cmfixed and was focus an con- centratorinner diametercapable ofof 4 concentration cm. This reaction factors tube was of connectedup to 2000 to suns. a 2 L reservoir A circulating flask, which reactor was positioned in the focalwas point cooled in(1–2 an ice cm bath in duringdiam operation.eter) of the The reactionsolar dish mixture. The was reactor purged with consisted oxygen positioned in the focal point (1–2 cm in diameter) of the solar dish. The reactor consisted of a double-jacked Pyrexin the reservoir tubing and of then30 cm circulated length through with the an inner outer reaction diameter tubing atof 500 5.5 mL/min, cm and while an inner diameter ofof 4a collingcm.double-jacked This water reaction was Pyrex passed tube tubing through was of the 30connected outercm leng mantle.th to with Aftera 2 an L six outerreservoir hours diameter of circulation flask, of 5.5which with cm and an inner diameter of 4 cm. This reaction tube was connected to a 2 L reservoir flask, which was cooled in an ice occasionalbath during refeeding operation. of methylene The blue,reaction the reaction mixture product was waspurged treated with with oxygen sodium wasmetabisulfite cooled in an (Na ice2 bathS2O5), during acidified operation. to pH 4–5 The and reaction the precipitated mixture oxygenated was purged rosin with col- oxygen in the reservoir inand the then reservoir circulated and then throug circulatedh the throug innerh reactionthe inner tubingreaction at tubing 500 mL/min, at 500 mL/min, while colling waterwhile was colling passed water through was passed the throughouter mantle. the outer After mantle. six Afterhours six of hours circulation of circulation with occasional withrefeeding occasional of methylene refeeding of blue methylene, the reaction blue, the product reaction was product treated was withtreated so- with sodium metabisulfitedium (Na metabisulfite2S2O5), acidified (Na2S2 Oto5), pH acidified 4–5 and to pH the 4–5 precipitated and the precipitated oxygenated oxygenated rosin rosin collected and analyzed.collected NMR and analyzed. and IR analysesNMR and revealedIR analyses the revealed presence the ofpresence photooxygenated of photooxygenated products with potentiallyproducts with enhanc potentiallyed emulsifier enhanced properties. emulsifier properties.As rosin consists As rosin mainlyconsists mainlyof ab- of abietic acid (39), its photooxygenation-reduction sequence to the oxygenated products 40– ietic acid (39), its photooxygenation-reduction sequence to the oxygenated products 40– 42 is exemplarily shown in Scheme 15. The author also suggested a potential industrial 42 is exemplarilyproduction shown in of Scheme photooxygenated 15. The author rosin from also agricultural suggested leftov a potentialer stocks industrial using four para- production of photooxygenatedbolic trough collectors rosin with from a moderate agricultural concen leftovtrationer factor stocks of using50–100 foursuns andpara- an aper- bolic trough collectors with a moderate concentration factor of 50–100 suns and an aper-

Molecules 2021, 26, x FOR PEER REVIEW 21 of 28 Molecules 2021, 26, 1685 20 of 27

2 ture oflected 10 m and each analyzed. connected NMR andin series. IR analyses In favorable revealed the regions presence with of photooxygenatedhigh levels of solar radiation,products the plant with was potentially estimated enhanced to convert emulsiﬁer 100 properties. kg of rosin As rosinin 300 consists L of mainlymethanol of in three hoursabietic in summer, acid (39 ),four its photooxygenation-reductionto five hours in spring and sequence six to eight to the hours oxygenated in winter, products respectively. Molecules 2021, 26, x FOR PEER REVIEW 21 of 28 40–42 is exemplarily shown in Scheme 15. The author also suggested a potential industrial production of photooxygenated rosin from agricultural leftover stocks using four parabolic trough collectors with a moderate concentration factor of 50–100 suns and an aperture of 2 ture10 m of each10 m connected2 each connected in series. in series. In favorable In favorable regions regions with with high high levels levels of solar of solar radiation, radiation,the plant the wasplant estimated was estimated to convert to convert 100 kg 100 of rosin kg of in rosin 300 L in of 300 methanol L of methanol in three hoursin three in hourssummer, in summer, four to ﬁve four hours to five in hours spring in and spring six and to eight six to hours eight in hours winter, in winter, respectively. respectively.

Scheme 15. Photooxygenation and thermal reduction of abietic acid (39), the major component in rosin, to its oxygenated products 40–42.

Scheme 15. PhotooxygenationPhotooxygenationSensitized and and thermal thermal dehydrogenation reduction reduction of of abietic abietic acid reactions ( 39),), the the major major of natural component component pine in in rosin, rosin,rosin to to andits its oxygenated oxygenated abietic acid (39) were products 40–42. furthermore studied by Icli and co-workers in concentrated sunlight (Scheme 16) [118]. The authors utilized a non-specified solar dish concentrator with a diameter of 1 m and a cooled tube-shapedSensitizedSensitized dehydrogenation dehydrogenation reactor placed reactions in its focalof natural point. pine Solutions rosin and abieticcontaining acid ( 39 rosin)) were were or 39, ben- furthermorefurthermore studiedstudied byby Icli Icli and and co-workers co-workers in in concentrated concentrated sunlight sunlight (Scheme (Scheme 16)[ 16)118 ].[118]. The zophenoneTheauthors authors utilized(Ph utilized2CO) a non-speciﬁed and a non-specified copper solar acetate solar dish concentratordish (Cu(OAc) concentrator with2; (CHa with diameter3CO a diameter2)2 ofCu) 1 m inof and 1benzene m acooled and a were ex- posedcooledtube-shaped to sunlight tube-shaped reactor under reactor placed nitrogen placed in its focal in itsfor point. focal up point. Solutionsto three Solutions containing hours. containing Each rosin orrosinexperimental39, or benzophe- 39, ben- run was stoppedzophenonenone eight (Ph2CO) times(Ph and2CO) for copper and aeration/deg copper acetate acetate (Cu(OAc)assing (Cu(OAc)2; to (CH regenerate32CO; (CH2)2Cu)3CO 2incupric)2Cu) benzene in (Cubenzene were2+) ions. exposed were Following ex- to this posedsunlight to undersunlight nitrogen under for nitrogen up to three for up hours. to three Each experimentalhours. Each experimental run was stopped run eightwas approach, abietic acid (39) was quantitatively converted2+ to dehydroabietic acid (43). The times for aeration/degassing to regenerate cupric (Cu ) ions. Following2+ this approach, authorsstopped concluded eight times that for aeration/degthe selectiveassing solar to regeneratedehydrogenation cupric (Cu to) ions. 43 mayFollowing have this commercial approach,abietic acid abietic (39) wasacid quantitatively(39) was quantitatively converted converted to dehydroabietic to dehydroabietic acid (43). acid The (43 authors). The potential,authorsconcluded although concluded that the the that selective reaction the sele solarctive scale dehydrogenation solar and dehydrogenation hazardous to 43 maysolvent to have 43 maynecessitate commercial have commercial potential,further optimization studies.potential,although thealthough reaction the scale reaction and hazardousscale and ha solventzardous necessitate solvent necessitate further optimization further optimiza- studies. tion studies.

Scheme 16. Photodehydrogenation of abietic acid (39) to dehydroabietic acid (43). SchemeScheme 16. Photodehydrogenation 16. Photodehydrogenation of of abietic acid acid (39 ()39 to) dehydroabieticto dehydroabietic acid (43 acid). (43). TheThe benzophenone-sensitized photodehydrogenation photodehydrogenation of of acenaphthene acenaphthene ( (4444)) in in the presenceThepresence benzophenone-sensitized of cupric pivalatepivalate (CuPiv(CuPiv22)) was was photodehydrogenation only only observed observed in in concentrated concentrated of sunlightacenaphthene sunlight [85 [85].]. Using Us- (44) in the presenceinga ﬁx a focusfixof focuscupric solar solar dish pivalate dish reactor reactor (CuPiv with with a sunlight2 )a wassunlight concentrationonly concentration observed factor factorin below concentrated below 15 suns 15 suns (Figure sunlight (Figure3), a [85]. Us- ing a3), degassedfix a focusdegassed solution solar solution dish of 44 ,ofreactor benzophenone 44, benzophenone with a and sunlight CuPiv and CuPiv2 inconcentration carbon2 in carbon tetrachloride tetrachloride factor was below illuminated was 15 illu- suns (Figure minatedfor 30 h overfor 30 a seven-dayh over a seven-day period with period several with aeration/degassing several aeration/degassing cycles. Subsequent cycles. Subse- GC– 3), a quentdegassedMS analysis GC–MS solution revealed analysis theof revealed 44 clean, benzophenone formationthe clean formation of acenaphthenone and of CuPiv acenaphthenone2 in (45 carbon) in 95%, (45 tetrachloride) next in 95%, to 5% next of was illuminated for 30 h over a seven-day period with several aeration/degassing cycles.1 Subse- tounreacted 5% of unreacted44 (Scheme 44 (Scheme 17). Continuous 17). Continuous solar exposure solar exposure under deaerated under deaerated conditions conditions for 7 2 h quentforwith GC–MS 7½ a concentrationhours analysis with a concentration factor revealed of under the factor 110 clean suns of under formation resulted 110 insuns aof low resultedacenaphthenone conversion in a low of just conversion 3%(45 with) in 95%, next to 5%ofacetnaphthylene of just unreacted 3% with acetnaphthylene 44 (46 (Scheme) detected 17). by (46 GC–MS )Continuous detected in anby amountGC–MS solar ofexposurein <1%.an amount Prolonged under of <1%. illuminationsdeaerated Prolonged conditions for 7½illuminationsfor hours up to 20with h for under aup concentration to the 20 same h under conditions the factor same somewhat of conditions under increased 110 somewhat suns conversion resulted increased rates inconversion a tolow 11%, conversion rates to 11%, but with only ca. 2% of 46 formed. The latter experiments unambiguously of just 3% with acetnaphthylene (46) detected by GC–MS in an amount of <1%. Prolonged showed the importance of molecular oxygen (3O2) for catalyst regeneration. illuminations for up to 20 h under the same conditions somewhat increased conversion rates to 11%, but with only ca. 2% of 46 formed. The latter experiments unambiguously showed the importance of molecular oxygen (3O2) for catalyst regeneration.

Molecules 2021, 26, 1685 21 of 27

Molecules 2021, 26, x FOR PEER REVIEW 22 of 28 but with only ca. 2% of 46 formed. The latter experiments unambiguously showed the 3 importance of molecular oxygen ( O2) for catalyst regeneration.

SchemeScheme 17. Photodehydrogenation 17. Photodehydrogenation of acenaphthene of acenaphthene (44) to (44 acenaphthenone) to acenaphthenone (45). (45).

Icli andIcli andco-workers co-workers also alsoinvestigated investigated the thenaphthalene naphthalene diimide-sensitized diimide-sensitized photooxi- photooxida- dationtion of ofanthracene anthracene to to its its corresponding corresponding endoperoxide endoperoxide in in the the HTC reactorreactor underunder 10.210.2 suns suns(Figure (Figure3 )[3) 119[119].]. However, However, no no further further experimental experimental details details were were provided provided by by the the authors. authors. 5. Limitations, Challenges and Opportunities 5. Limitations,Although Challenges photooxygenations and Opportunities have been successfully realized ‘outdoors’ in natural sunlight on the demonstration to technical scales, many reported processes suffer from Although photooxygenations have been successfully realized ‘outdoors’ in natural experimental or technical drawbacks. For example, very few solar manufacturing stud- sunlight on the demonstration to technical scales, many reported processes suffer from ies, notably rose oxide (15)[59], (-)-trans-4-hydroxy-β-thujene (23a)[102] and juglone experimental or technical drawbacks. For example, very few solar manufacturing studies, (31)[91,114,115], have truly isolated the oxygenated target products of interest. Most notably rose oxide (15) [59], (-)-trans-4-hydroxy-β-thujene (23a) [102] and juglone (31) investigations obtained crude products instead and determined compositions and ‘yields’ [91,114,115], have truly isolated the oxygenated target products of interest. Most investi- simply by NMR, GC or GC–MS analyses. A further conversion of a crude reaction mix- gations obtained crude products instead and determined compositions and ‘yields’ ture followed by isolation of the desired product (hexyloxyfuranone 28; R = Hex) has simplynevertheless by NMR, GC been or GC–MS described analyses. [111]. SomeA further processes conversion generated of a crude complex reaction mixtures, mixture hence followedrequiring by isolation time and of resourcethe desired demanding product (hexyloxyfuranone separation and purification 28; R = Hex) steps. has Several never- pro- thelesstocols been also described involved [111]. hazardous Some processe and flammables generated solvents complex from mixtures, the corresponding hence requir- ‘indoor’ ing time(laboratory) and resource procedure, demanding which separation need major and risk purification control measures. steps. Several Other protocols studies also utilized involvedhighly hazardous sunlight-concentrating and flammable facilitiessolvents fr suchom the as parabolic corresponding dish systems ‘indoor’ or(laboratory) solar furnaces procedure,that place which the need reaction major chamber risk control within measures. their focal Other points. studies Next utilized to the highhighly installation sunlight-concentratingcosts, these concentrators facilities such generate as paraboli large amountsc dish systems of heat or and solar thus furnaces demand that sophisticated place the reactionand reliable chamber cooling within systems their for focal effective points. heat Next management. to the high installation Due to the limitedcosts, these capacity concentratorsof the actual generate reaction large chambers, amounts of these heat systems and thus are demand also difficult sophisticated to scale and to reliable industrially coolingrelevant systems productivities. for effective heat management. Due to the limited capacity of the actual reaction chambers,Preparative these solar systems chemistry are hasalso seen diffic ault surge to scale between to industrially 1990 and 2005, relevant largely produc- driven by tivities.dedicated research programs involving the PSA in Spain and the DLR in Germany [120]. DespitePreparative the realizationsolar chemistry of a variety has seen of a attractive surge between solar transformations 1990 and 2005, duringlargely thisdriven period, by dedicatedthe developed research technologies programs and involving processes the were PSA not in implementedSpain and the by DLR the chemicalin Germany industry. [120].A Despite major hurdle the realization for adaptation of a variety may have of attractive been the highsolar initial transformations investment during costs, especially this period,for the more developed advanced technologies sunlight concentrator and processes systems, were not and implemented the need for by ‘operation the chemical security’ industry.independent A major of hurdle location for or adaptation weather. In may contrast have tobeen the the more high complex initial troughinvestment or dish costs, reactors, especiallylow sunlight-concentrating for more advanced sunlight systems concen suchtrator as the systems, CPC reactor and are the easy need to for manufacture ‘operation and security’maintain independent and can furthermoreof location or utilize weather. direct In andcontrast diffuse to radiation.the more complex This makes trough this or device dish advantageousreactors, low forsunlight-concentrating the construction of solar systems chemical such plants, as the as CPC already reactor successfully are easy demon-to manufacturestrated for and large-scale maintain and wastewater can furthermore treatment utilize processes direct [ 57and,58 diffuse]. The initialradiation. construction This makescosts this for device these advantageous facilities may for be the offset constr byuction annual of savings solar chemical on energy, plants, cooling as already water and successfullyreplacement demonstrated lamps [95 ].for At large-scale present, the wast transferewater of treatment existing photochemicalprocesses [57,58]. processes The in- from itial constructionthe fragrance costs and flavorfor these industry facilities to solarmay be operations offset by appears annual savings economically on energy, and technically cooling waterviable. and These replacement (and other) lamps fine and [95]. specialty At presen productst, the transfer are manufactured of existing onphotochemical scales below 1000 processestons/year from andthe fragrance thus do not and necessarily flavor indust requirery to continuous, solar operations year-round appears operation economically [121– 123]. and technicallyNaturally, viable. the These location (and of other) any futurefine and synthetic-organic specialty products solar are industry manufactured will demand on scalesfavorable below 1000 sunlight tons/year conditions, and thus but do also no existingt necessarily infrastructure require continuous, and access to year-round target markets. operation [121–123]. Naturally, the location of any future synthetic-organic solar industry will demand favorable sunlight conditions, but also existing infrastructure and access to target markets.

Molecules 2021, 26, 1685 22 of 27

For larger chemical markets or less favorable locations, however, tandem solar lamp-driven processes can overcome the reliance on the day–night cycle and hence guarantee a continuous production of target materials. Under inadequate sunlight conditions or at night, energy efﬁcient light sources such as LEDs or OLEDs can drive or support the photochemical process [13,124,125]. Prototype solar lamp-driven devices have indeed already been developed and applied for water treatment applications [126,127]. Solar photochemistry also offers economic beneﬁts for decentralized manufacturing, for example small-scale conversions of local biomass in the (tropical) essential oil industry or in developing countries [128,129]. In isolated and remote locations, photovoltaic (PV) panels can provide power for electrical components, as again successfully demonstrated for water treatment applications [130,131].

6. Conclusions and Outlook Photooxygenations are ideal candidates for preparative solar applications with signiﬁ- cant future synthesis potential [132,133]. Industrially relevant and new processes have been realized using a variety of solar reactor systems, clearly demonstrating the capability of this methodology. Current or future technologies and processes developed can be further adopted and transferred to other photochemical or even thermal applications [39–47,134]. Visible-light photoredox catalysis, for example, has seen remarkable growth over the last decade [135–137], but no large-scale solar processes have been reported yet [138]. Likewise, new solar chemical reactor concepts have been recently developed, including continuous-ﬂow trough and luminescent solar concentrators [139–141]. These chemical and technological advances have complemented past and existing activities in green photochemistry [142–148]. It is hoped that the future of solar photochemistry in general as a sustainable synthesis tool will be bright [149].

Author Contributions: J.S.W. wrote the original draft manuscript; M.J.R. provided corrections to the final version; M.O. revised the manuscript and secured the funding. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by the Australian Research Council (ARC; grant number DP130100794) and the Queensland Department of Employment, Economic Development & In- novation (DEEDI; Proof of Concept Grant 2012). Acknowledgments: The authors thank emeritus Prof Jochen Mattay, Christian Jung, Christian Sattler, Jürgen Ortner and Peter Esser for their ongoing support over the years. The authors also thank emeritus Prof Siddik Icli for information on the parabolic dish solar concentrator. Conflicts of Interest: The authors declare no conflict of interest.

References 1. Bonﬁeld, H.E.; Knauber, T.; Lévesque, F.; Moschetta, E.G.; Susanne, F.; Edwards, L.J. Photons as a 21st century reagent. Nat. Commun. 2020, 11, 1–4. [CrossRef][PubMed] 2. Griesbeck, A.G. The Future of Photochemistry: Just Bright. ChemPhotoChem 2018, 3, 8–9. [CrossRef] 3. Pagliaro, M. Making APIs and ﬁne chemicals with light. Chim. Oggi—Chem. Today 2017, 35, 84–85. 4. Bach, T. More Chemistry with light! More light in chemistry! Angew. Chem. Int. Ed. 2015, 54, 11294–11295. [CrossRef] 5. Balzani, V.; Bergamini, G.; Ceroni, P. Photochemistry and photocatalysis. Rend. Lincei 2017, 28, 125–142. [CrossRef] 6. Kärkäs, M.D.; Porco, J.A., Jr.; Stephenson, C.R.J. Photochemical approaches to complex chemotypes: Applications in natural product synthesis. Chem. Rev. 2016, 116, 9683–9747. [CrossRef] 7. Oelgemöller, M.; Hoffmann, N. Studies in organic and physical photochemistry—An interdisciplinary approach. Org. Biomol. Chem. 2016, 14, 7392–7442. [CrossRef][PubMed] 8. Hoffmann, N. Photochemical Reactions as Key Steps in Organic Synthesis. Chem. Rev. 2008, 108, 1052–1103. [CrossRef] 9. Rehm, T.H. Reactor Technology Concepts for Flow Photochemistry. ChemPhotoChem 2019, 4, 235–254. [CrossRef] 10. Di Filippo, M.; Bracken, C.; Baumann, M. Continuous Flow Photochemistry for the Preparation of Bioactive Molecules. Molecules 2020, 25, 356. [CrossRef] 11. Politano, F.; Oksdath-Mansilla, G. Light on the Horizon: Current Research and Future Perspectives in Flow Photochemistry. Org. Process. Res. Dev. 2018, 22, 1045–1062. [CrossRef] Molecules 2021, 26, 1685 23 of 27

12. Oelgemöller, M.; Hoffmann, N.; Shvydkiv, O.; Oelgemoeller, M. From ‘Lab & Light on a Chip’ to Parallel Microflow Photochem- istry. Aust. J. Chem. 2014, 67, 337–342. [CrossRef] 13. Sender, M.; Ziegenbalg, D. Light Sources for Photochemical Processes—Estimation of Technological Potentials. Chem. Ing. Tech. 2017, 89, 1159–1173. [CrossRef] 14. Braun, A.M.; Oller do Nascimento, C.A. Angewandte (präparative) Photochemie. Nachr. Chem. Tech. Lab. 1991, 39, 515–526. 15. Braun, A.; Peschl, G.H.; Oliveros, E. Industrial photochemistry. In CRC Handbook of Organic Photochemistry and Photobiology, 3rd ed.; Griesbeck, A., Oelgemöller, M., Ghetti, F., Eds.; CRC Press: Boca Raton, FL, USA, 2012; Volume 1, Chapter 1; pp. 1–19. 16. Michelin, C.; Lefebvre, C.; Hoffmann, N. Les réactions photochimiques à l’échelle industrielle. L’Act. Chim. 2019, 436, 19–27. 17. Pfoertner, K.-H.; Oppenländer, T. Photochemistry. In Ullmann’s Encyclopedia of Industrial Chemistry; Wiley-VCH: Weinheim, Germany, 2012; pp. 1–45. 18. Braun, A.M.; Jakob, L.; Oliveros, E.; Nascimento, C.A.O.D. Up-Scaling Photochemical Reactions. Adv. Photochem. 2007, 18, 235–313. [CrossRef] 19. Fischer, M. Industrial Applications of Photochemical Syntheses. Angew. Chem. Int. Ed. 1978, 17, 16–26. [CrossRef] 20. Mattay, J. Von der Laborsynthese zur Solarchemie. Chem. unsere Zeit 2002, 36, 98–106. [CrossRef] 21. Wilkins, F.W.; Blake, D.M. Use solar energy to drive chemical processes. Chem. Eng. Prog. 1994, 90, 41–49. 22. Funken, K.-H. Solare Photoreaktionen für die chemische Technik. Nachr. Chem. Tech. Lab. 1992, 40, 793–800. [CrossRef] 23. Mathur, V.; Wong, E. Production of fuels and chemicals using solar photothermochemistry. Energy 1987, 12, 311–318. [CrossRef] 24. Albini, A.; Dichiarante, V. The ‘belle ’epoque’ of photochemistry. Photochem. Photobiol. Sci. 2009, 8, 248–254. [CrossRef] 25. Roth, H.D. Twentieth century developments in photochemistry. Brief historical sketches. Pure Appl. Chem. 2001, 73, 395–403. [CrossRef] 26. Roth, H.D. The Beginnings of Organic Photochemistry. Angew. Chem. Int. Ed. 1989, 28, 1193–1207. [CrossRef] 27. Ciamician, G. The photochemistry of the future. Science 1912, 36, 385–394. [CrossRef] 28. Protti, S.; Fagnoni, M. The sunny side of chemistry: Green synthesis by solar light. Photochem. Photobiol. Sci. 2009, 8, 1499–1516. [CrossRef] 29. Ravelli, D.; Protti, S.; Fagnoni, M. Application of Visible and Solar Light in Organic Synthesis. Lect. Notes Chem. 2016, 92, 281–342. [CrossRef] 30. Fuqiang, W.; Ziming, C.; Jianyu, T.; Yuan, Y.; Yong, S.; Linhua, L. Progress in concentrated solar power technology with parabolic trough collector system: A comprehensive review. Renew. Sustain. Energy Rev. 2017, 79, 1314–1328. [CrossRef] 31. Weinstein, L.A.; Loomis, J.; Bhatia, B.S.; Bierman, D.M.; Wang, E.N.; Chen, G. Concentrating Solar Power. Chem. Rev. 2015, 115, 12797–12838. [CrossRef] 32. Fernández-García, A.; Zarza, E.; Valenzuela, L.; Pérez, M. Parabolic-trough solar collectors and their applications. Renew. Sustain. Energy Rev. 2010, 14, 1695–1721. [CrossRef] 33. Roeb, M.; Neises, M.; Monnerie, N.; Sattler, C.; Pitz-Paal, R. Technologies and trends in solar power and fuels. Energy Environ. Sci. 2011, 4, 2503–2511. [CrossRef] 34. Funken, K.-H.; Ortner, J. Technologies for the Solar Photochemical and Photocatalytic Manufacture of Specialities and Commodi- ties: A Review. Z. Phys. Chem. 1999, 213, 99–105. [CrossRef] 35. Fendrich, M.A.; Quaranta, A.; Orlandi, M.; Bettonte, M.; Miotello, A. Solar Concentration for Wastewaters Remediation: A Review of Materials and Technologies. Appl. Sci. 2018, 9, 118. [CrossRef] 36. Malato, S.; Maldonado, M.I.; Fernández-Ibáñez, P.; Oller, I.; Polo, I.; Sánchez-Moreno, R. Decontamination and disinfection of water by solar photocatalysis: The pilot plants of the Plataforma Solar de Almeria. Mater. Sci. Semicond. Process. 2016, 42, 15–23. [CrossRef] 37. Tanveer, M.; Guyer, G.T. Solar assisted photo degradation of wastewater by compound parabolic collectors: Review of design and operational parameters. Renew. Sustain. Energy Rev. 2013, 24, 534–543. [CrossRef] 38. Bahnemann, D. Photocatalytic water treatment: Solar energy applications. Sol. Energy 2004, 77, 445–459. [CrossRef] 39. Oelgemöller, M. Solar Photochemical Synthesis: From the Beginnings of Organic Photochemistry to the Solar Manufacturing of Commodity Chemicals. Chem. Rev. 2016, 116, 9664–9682. [CrossRef] 40. Spasiano, D.; Marotta, R.; Malato, S.; Fernandez-Ibañez, P. Solar photocatalysis: Materials, reactors, some commercial, and pre-industrialized applications. A comprehensive approach. Appl. Cat. B Environ. 2015, 170, 90–123. [CrossRef] 41. Oelgemöller, M.; Jung, C.; Mattay, J.; Oelgemoeller, M. Green photochemistry: Production of fine chemicals with sunlight. Pure Appl. Chem. 2007, 79, 1939–1947. [CrossRef] 42. Esser, P.; Pohlmann, B.; Scharf, H.-D. The Photochemical Synthesis of Fine Chemicals with Sunlight. Angew. Chem. Int. Ed. 1994, 33, 2009–2023. [CrossRef] 43. Mumtaz, S.; Sattler, C.; Oelgemöller, M. Solar photochemical manufacturing of fine chemicals—historical background, modern solar technologies, recent applications and future challenges. In Chemical Processes for a Sustainable Future; Letcher, T.M., Scott, J.L., Patterson, D., Eds.; Royal Society of Chemistry: Cambridge, UK, 2015; Chapter 7; pp. 158–191. 44. Coyle, E.E.; Oelgemöller, M. Solar photochemistry from the beginnings of organic photochemistry to the solar production of chem-icals. In CRC Handbook of Organic Photochemistry and Photobiology, 3rd ed.; Griesbeck, A., Oelgemöller, M., Ghetti, F., Eds.; CRC Press: Boca Raton, FL, USA, 2012; Volume 1, Chapter 10; pp. 237–247. Molecules 2021, 26, 1685 24 of 27

45. Oelgemöller, M.; Jung, C.; Ortner, J.; Mattay, J.; Schiel, C.; Zimmermann, E. Green photochemistry with moderately concentrated sunlight. Spectrum 2005, 18, 28–33. 46. Oelgemöller, M.; Mattay, J.; Jung, C.; Schiel, C.; Ortner, C.; Zimmermann, E. Back to the roofs—the solarchemical production of fine chemicals with sunlight. In Proceedings of the International Solar Energy Conference, ISEC 2004 Conference, Portland, OR, USA, 11–14 July 2004; American Society of Mechanical Engineers (ASME): New York, NY, USA, 2004; pp. 523–531. 47. Pohlmann, B.; Scharf, H.-D.; Jarolimek, U.; Mauermann, P. Photochemical production of fine chemicals with concentrated sunlight. Sol. Energy 1997, 61, 159–168. [CrossRef] 48. Fu, Q. Radiation (solar). In Encyclopedia of Atmospheric Sciences; Holton, J., Ed.; Academic Press: Amsterdam, The Netherlands, 2003; pp. 1859–1863. 49. Iqbal, M. An Introduction to Solar Radiation; Academic Press: New York, NY, USA, 1983. 50. Pibiri, I.B.; Piccionello, S.; Palumbo, A.; Pace, A. Photochemically produced singlet oxygen: Applications and perspectives. ChemPhotoChem 2018, 2, 535–547. [CrossRef] 51. Ghogare, A.A.; Greer, A. Using Singlet Oxygen to Synthesize Natural Products and Drugs. Chem. Rev. 2016, 116, 9994–10034. [CrossRef] 52. DeRosa, M.C.; Crutchley, R.J. Photosensitized singlet oxygen and its applications. Coord. Chem. Rev. 2002, 233–234, 351–371. [CrossRef] 53. Turconi, J.; Griolet, F.; Guevel, R.; Oddon, G.; Villa, R.; Geatti, A.; Hvala, M.; Rossen, K.; Göller, R.; Burgard, A. Semisynthetic Artemisinin, the Chemical Path to Industrial Production. Org. Process. Res. Dev. 2014, 18, 417–422. [CrossRef] 54. Gollnick, K. Photooxygenation and its application in industry. Chim. Ind. 1982, 64, 156–166. [CrossRef] 55. Rojahn, W.; Warnecke, H.-U. Die Photosensibilisierte Sauerstoffübertragung—eine Methode zur Herstellung hochwertiger Riech-stoffe. Dragoc. Rep. 1980, 27, 159–164. 56. Demuth, M. Chemie mit Sonnenlicht—auch bei Bewölkung. In Innovationspreis Ruhrgebiet; Kommunalverband Ruhrgebiet: Essen, Germany, 2000; pp. 91–94. 57. Blanco, J.; Malato, S.; Fernández, P.; Vidal, A.; Morales, A.; Trincado, P.; Oliveira, J.; Minero, C.; Musci, M.; Casalle, C.; et al. Compound parabolic concentrator technology development to commercial solar detoxification applications. Sol. Energy 1999, 67, 317–330. [CrossRef] 58. Ajona, J.; Vidal, A. The use of CPC collectors for detoxification of contaminated water: Design, construction and preliminary results. Sol. Energy 2000, 68, 109–120. [CrossRef] 59. Jung, C.; Funken, K.-H.; Ortner, J. Prophis: Parabolic trough-facility for organic photochemical syntheses in sunlight. Photochem. Photobiol. Sci. 2005, 4, 409–411. [CrossRef] 60. Funken, K.-H. Solar chemistry using concentrating solar facilities. In Proceedings of the Sixth International Summer School Solar Energy 2000, Klagenfurt, Austria, 24 July–4 August 2000; Faninger, G., Bucher, W., Wolfer, U., Eds.; Interuniversitäres Institut für interdisziplinäre Forschung und Fortbildung (IFF), Universität Klagenfurt: Klagenfurt, Austria, 2001; pp. 118–137. 61. Neumann, A.; Groer, U. Experimenting with concentrated sunlight using the DLR solar furnace. Sol. Energy 1996, 58, 181–190. [CrossRef] 62. Fields, C.L.; Pitts, J.R.; Hale, M.J.; Bingham, C.; Lewandowski, A.; King, D.E. Formation of fullerenes in highly concentrated solar flux. J. Phys. Chem. 1993, 97, 8701–8702. [CrossRef] 63. Flamant, G.; Luxembourg, D.; Robert, J.; Laplaze, D. Optimizing fullerene synthesis in a 50 kW solar reactor. Sol. Energy 2004, 77, 73–80. [CrossRef] 64. Foote, C.S. Definition of type I and type II photosensitized oxidation. Photochem. Photobiol. 1991, 54, 659. [CrossRef] 65. Schmidt, R. Photosensitized Generation of Singlet Oxygen. Photochem. Photobiol. 2007, 82, 1161–1177. [CrossRef] 66. Osterberg, P.M.; Niemeier, J.K.; Welch, C.J.; Hawkins, J.M.; Martinelli, J.R.; Johnson, T.E.; Root, T.W.; Stahl, S.S. Experimental lim-iting oxygen concentrations for nine organic solvents at temperatures and pressures relevant to aerobic oxidations in the pharma-ceutical industry. Org. Process. Res. Dev. 2015, 19, 1537–1543. [CrossRef][PubMed] 67. Montalti, M.; Credi, A.; Prodi, L.; Gandolfi, M.T. Handbook of Photochemistry, 3rd ed.; CRC Press: Boca Raton, FL, USA, 2006; pp. 542–548. 68. Clarke, C.J.; Tu, W.-C.; Levers, O.; Bröhl, A.; Hallett, J.P. Green and Sustainable Solvents in Chemical Processes. Chem. Rev. 2018, 118, 747–800. [CrossRef] 69. Fresnadillo, D.G.; Lacombe, S. Reference photosensitizers for the production of singlet oxygen. In Singlet Oxygen: Applications in Biosciences and Nanosciences; Nonell, S., Flors, C., Eds.; The Royal Society of Chemistry: Cambridge, UK, 2016; Volume 1, Chapter 6; pp. 105–143. 70. Marin, M.L.; Santos-Juanes, L.; Arques, A.; Amat, A.M.; Miranda, M.A. Organic photocatalysts for the oxidation of pollutants and model compounds. Chem. Rev. 2012, 112, 1710–1750. [CrossRef] 71. Lacombe, S.; Pigot, T. Materials for selective photo-oxygenation vs. photocatalysis: Preparation, properties and applications in en-vironmental and health fields. Catal. Sci. Technol. 2016, 6, 1571–1592. [CrossRef] 72. Wahlen, J.; De Vos, D.E.; Jacobs, P.A.; Alsters, P.L. Solid Materials as Sources for Synthetically Useful Singlet Oxygen. Adv. Synth. Catal. 2004, 346, 152–164. [CrossRef] Molecules 2021, 26, 1685 25 of 27

73. Malakar, P.; Deb, A.R.; Goodine, T.; Robertson, M.J.; Oelgemöller, M. Continuous-flow photooxygenations: an advantageous and sustainable oxidation methodology with a bright future. In Catalytic Aerobic Oxidations—Catalysis Series; Mejía, E., Ed.; Royal Society of Chemistry: Cambridge, UK, 2020; Chapter 7, pp. 181–251. [CrossRef] 74. Nardello-Rataj, V.; Alsters, P.L.; Aubry, J.-M. Industrial Prospects for the Chemical and Photochemical Singlet Oxygenation of Organic Compounds. Liq. Phase Aerob. Oxid. Catal. Ind. Appl. Acad. Persp. 2016, 22, 369–395. [CrossRef] 75. Alberti, M.N.; Orfanopoulos, M. Unravelling the mechanism of the singlet oxygen ene reaction: Recent computational and experi-mental approaches. Chem. Eur. J. 2010, 16, 9414–9421. [CrossRef][PubMed] 76. Bayer, P.; Pérez-Ruiz, R.; Von Wangelin, A.J. Stereoselective Photooxidations by the Schenck Ene Reaction. ChemPhotoChem 2018, 2, 559–570. [CrossRef] 77. Khayyat, S.A.; Roselin, L.S. Recent progress in photochemical reaction on main components of some essential oils. J. Saudi Chem. Soc. 2018, 22, 855–875. [CrossRef] 78. Goodine, T.; Oelgemöller, M. Corymbia citriodora: A Valuable Resource from Australian Flora for the Production of Fragrances, Repellents, and Bioactive Compounds. ChemBioEng Rev. 2020, 7, 170–192. [CrossRef] 79. Schenck, O.; Ziegler, K. Die Synthese des Ascaridols. Naturwissenschaften 1944, 32, 157. [CrossRef] 80. Schenck, G.O. Autoxydation von Furan und anderen Dienen (Die Synthese des Ascaridols). Ang. Chem. 1944, 57, 101–102. [CrossRef] 81. Schenck, G.; Schulze-Buschoff, H. Synthetisches Askaridol, eine neue Möglichkeit der Spulwurmbehandlung. DMW—Dtsch. Med. Wochenschr. 1948, 73, 341–344. [CrossRef][PubMed] 82. Karapire, C.; Kolancilar, H.; Oyman, Ü.; Icli, S. Fluorescence emission and photooxidation studies with 5,6- and 6,7- benzocoumarins and a 5,6-benzochromone under direct and concentrated sun light. J. Photochem. Photobiol. Chem. 2002, 153, 173–184. [CrossRef] 83. Dindar, B.; Içli, S. Unusual photoreactivity of zinc oxide irradiated by concentrated sunlight. J. Photochem. Photobiol. Chem. 2001, 140, 263–268. [CrossRef] 84. Karapire, C.; Ku´s,M.; Turkmen, G.; Trevithick-Sutton, C.; Foote, C.; Icli, S. Photooxidation studies with perylenediimides in solution, PVC and sol–gel thin films under concentrated sun light. Sol. Energy 2005, 78, 5–17. [CrossRef] 85. Avcibasi, N.; Icli, S.; Gilbert, A. Photochemical reactions of α-terpinene and acenaphthene under concentrated sunlight. Turk. J. Chem. 2003, 27, 1–7. 86. Ohloff, G.; Klein, E.; Schenck, G.O. Darstellung von “Rosenoxyden” und anderen Hydropyran-Derivaten über Photohydroper- oxyde. Angew. Chem. 1961, 73, 578. [CrossRef] 87. Bicas, J.L.; Dionísio, A.P.; Pastore, G.M. Bio-oxidation of Terpenes: An Approach for the Flavor Industry. Chem. Rev. 2009, 109, 4518–4531. [CrossRef] 88. Pickenhagen, W.; Schatkowski, D. Verfahren zur Herstellung von Rosenoxid. DE19645922A1, 14 February 2002. 89. Demuth, M.; Ritter, A. Photochemical and Thermochemical Solar Syntheses Using Flat-Bed Solar Collectors/Solar Reactors. WO99/54042A1, 28 October 1999. 90. Ritterskamp, P.; Hülsdünker, A.; Heimann, F.; Goeller, F.; Uzun, D.; Ritter, A.; Demuth, M.; Kleinwächter, J.; Ortner, J.; Funken, K.-H.; et al. Synthesis of biologically active compounds with sunlight and technological developments. In Solare Chemie und Solare Materialforschung; AG-Solar NRW: Jülich, Germany, 2002; Chapter 2.2.1; pp. 1–6. ISBN 3893363068. 91. Oelgemöller, M.; Jung, C.; Ortner, J.; Mattay, J.; Zimmermann, E.; Oelgemoeller, M. Green photochemistry: Solar photooxygenations with medium concentrated sunlight. Green Chem. 2004, 7, 35–38. [CrossRef] 92. Ortner, J.; Faust, D.; Funken, K.-H.; Lindner, T.; Schulat, J.; Stojanoff, C.G.; Fröning, P. New developments using holographic concentration in solar photochemical reactors. J. Phys. Colloq. 1999, 9, Pr3-379–Pr3-383. [CrossRef] 93. Stojanoff, C.G.; Schulat, J.; Fröning, P. Development and fabrication of holographic concentrators for solar chemistry. In Solare Chemie und Solare Materialforschung; AG-Solar NRW: Jülich, Germany, 2002; Chapter 6.4; pp. 1–10. ISBN 3893363068. 94. Funken, K.-H.; Pohlmann, B.; Ortner, J.; Faust, D. Verfahren zur Oxidation ungesättigter Kohlenwasserstoffe. DE19923071A1, 21 August 2003. 95. Monniere, N.; Ortner, J. Economic evaluation of the industrial photosynthesis of rose oxide via lamp or solar operated photooxidation of citronellol. J. Sol. Energy Eng. 2001, 123, 171–174. [CrossRef] 96. Dincalp, H.; Icli, S. Photosynthesis of rose oxide by concentrated sunlight in the absence of singlet oxygen. J. Photochem. Photobiol. Chem. 2001, 141, 147–151. [CrossRef] 97. Schenck, G.O.; Eggert, H.; Denk, W. Über die Bildung von Hydroperoxyden bei photosensibilisierten Reaktionen von O2 mit ge-eigneten Akzeptoren, insbesondere mit α- und β-Pinen. Liebigs Ann. Chem. 1953, 584, 177–198. [CrossRef] 98. Bhatia, S.; McGinty, D.; Letizia, C.; Api, A. Fragrance material review on myrtenol. Food Chem. Toxicol. 2008, 46, S237–S240. [CrossRef][PubMed] 99. Mihelich, E.D.; Eickhoff, D.J. A one-pot conversion of olefins to. alpha.,. beta.-unsaturated carbonyl compounds. An easy synthesis of 2-cyclopentenone and related compounds. J. Org. Chem. 1983, 48, 4135–4137. [CrossRef] 100. Jung, C.; Ortner, J.; Jarolimek, U.; Mauermann, P.; Leyen, B. Advances in photochemical production of fine chemicals with con-centrated sunlight. In Solare Chemie und Solare Materialforschung; AG-Solar NRW: Jülich, Germany, 2002; Chapter 2.2.1; pp. 1–10. ISBN 3893363068. 101. Neckers, D.C. Rose Bengal. J. Photochem. Photobiol. Chem. 1989, 47, 1–29. [CrossRef] Molecules 2021, 26, 1685 26 of 27

102. Esser, P. Die Anwendung von Sonnenlicht zur Photochemischen Producktion von Feinchemikalien. Ph.D. Thesis, Rheinisch- Westfälische Technische Hochschule, Aachen, Germany, 1994. 103. Funken, K.-H.; Schneider, G.; Esser, E.P.; Scharf, H.-D.; Esser, P.; Wöhrle, I. The SOLARIS-experiment: Demonstration of solar-photochemical syntheses of fine chemicals. In Proceedings of the 6th International Symposium on Solar Thermal Concen- trating Technologies, Mojacar, Spain, 28 September–2 October 1992; Centro de Investigaciones Energétical, Medioambientales y Tecnológias (CIEMAT): Madrid, Spain, 1992; Volume 2, pp. 1027–1037. 104. Klein, E.; Rojahn, W. Die photosensibilisierte O2-Übertragung auf (+)-α-Thujen. Chem. Ber. 1965, 98, 3045–3049. [CrossRef] 105. Baeckström, P.; Koutek, B.; Šaman, D.; Vrkoˇc,J. A convenient synthesis of trans-sabinene hydrate from (−)-3-thujol via a highly selective ene reaction of singlet oxygen. Bioorg. Med. Chem. 1996, 4, 419–421. [CrossRef] 106. Scharf, H.-D.; Esser, P.; Kuhn, W.; Pelzer, R. Process for the Photooxidation of Terpene Olefins. US5620569A, 15 April 1997. 107. Schenck, G.O. Über die unsensibilisierte und photosensibilisierte Autoxydation von Furanen. Liebigs Ann. Chem. 1953, 584, 156–176. [CrossRef] 108. Schenck, G.O. Verfahren zur Herstellung von β-Acyl-acrylsäuren bzw. ihren Pseudoestern. DE875650C, 4 May 1953. 109. Bolz, G.; Wiersdorff, W.-W. Verfahren zur Herstellung von 2-Hydroxy-2,5-dihydrofuranon-(5). DE2111119A1, 14 September 1972. 110. Hoydonckx, H.E.; Van Rhijn, W.M.; De Vos, D.E.; A Jacobs, P. Furfural and Derivatives. In Ullmann’s Encyclopedia of Industrial Chemistry; Wiley: Hoboken, NJ, USA, 2007; Volume 16, pp. 285–313. 111. Esser, P.; Pelzer, R.; Völkl, F. Verwendung substituierter Lactone als Riechstoffe. EP0761808A2, 19 May 1999. 112. Strugstad, M.P.; Despotovski, S. A summary of extraction, synthesis, properties, and potential uses of juglone: A literature review. J. Ecosyst. Manag. 2012, 13, 1–16. 113. Griffiths, J.; Chu, K.-Y.; Hawkins, C. Photosensitised oxidation of 1-naphthols. J. Chem. Soc. Chem. Commun. 1976, 676–677. [CrossRef] 114. Oelgemöller, M.; Healy, N.; De Oliveira, L.; Jung, C.; Mattay, J. Green photochemistry: Solar-chemical synthesis of Juglone with medium concentrated sunlight. Green Chem. 2006, 8, 831–834. [CrossRef] 115. Suchard, O.; Kane, R.; Roe, B.J.; Zimmermann, E.; Jung, C.; Waske, P.A.; Mattay, J.; Oelgemöller, M. Photooxygenations of 1-naphthols: An environmentally friendly access to 1,4-naphthoquinones. Tetrahedron 2006, 62, 1467–1473. [CrossRef] 116. Dincalp, H.; Icli, S. Photoinduced electron transfer-catalyzed processes of sulfoamine perylene diimide under concentrated sun light. Sol. Energy 2006, 80, 332–346. [CrossRef] 117. Icli, S. Production of oxygenated rosin emulsifier by a solar photoorganic chemical method. For. Chem. Rev. 1999, 102, 7–9. 118. Icli, S.; Bulut, A.; Gül, Y. Room temperature generation of dehydrogenated colophony: Solar chemical production. Turk. J. Chem. 1992, 16, 289–292. 119. Erten, S.; Alp, S.; Icli, S. Photooxidation quantum yield efficiencies of naphthalene diimides under concentrated sun light in com-parisons with perylene diimides. J. Photochem. Photobiol. Chem. 2005, 175, 214–220. [CrossRef] 120. Funken, K.-H.; Ortner, J.; Riffelmann, K.-J.; Sattler, C. New developments in solar photochemistry. J. Inf. Rec. 1998, 24, 61–68. 121. Tirronen, E.; Salmi, T. Process development in the fine chemical industry. Chem. Eng. J. 2003, 91, 103–114. [CrossRef] 122. Bruggink, A. Growth and efficiency in the (fine) chemical industry. Chim. Oggi 1998, 16, 44–47. 123. Pollak, P.; Vouillamoz, R. Fine Chemicals. In Ullmann’s Encyclopedia of Industrial Chemistry; Wiley: Hoboken, NJ, USA, 2000; pp. 1–17. 124. Landgraf, S. Application of semiconductor light sources for investigations of photochemical reactions. Spectrochim. Acta Part A Mol. Biomol. Spectrosc. 2001, 57, 2029–2048. [CrossRef] 125. Kalyani, N.T.; Dhoble, S. Organic light emitting diodes: Energy saving lighting technology—A review. Renew. Sustain. Energy Rev. 2012, 16, 2696–2723. [CrossRef] 126. Monteiro, R.A.R.; Rodrigues-Silva, C.; Lopes, F.V.S.; Silva, A.M.T.; Boaventura, R.A.R.; Vilar, V.J.P. Evaluation of a solar/UV an-nular pilot scale reactor for 24 h continuous photocatalytic oxidation of n-decane. Chem. Eng. J. 2015, 280, 409–416. [CrossRef] 127. Navntoft, C.; Araujo, P.; Litter, M.I.; Apella, M.C.; Fernández, D.; Puchulu, M.E.; Hidalgo, M.D.V.; Blesa, M.A. Field Tests of the Solar Water Detoxification SOLWATER Reactor in Los Pereyra, Tucumán, Argentina. J. Sol. Energy Eng. 2006, 129, 127–134. [CrossRef] 128. Bolte, M.; Klaeden, K.; Beqiraj, A.; Oelgemöller, M. Photochemistry Down under—Solar chemicals from and for the Tropics. Eur. Photochem. Assoc. Newslett. 2013, 83–84, 79–83. 129. Gallezot, P. Conversion of biomass to selected chemical products. Chem. Soc. Rev. 2012, 41, 1538–1558. [CrossRef] 130. Horikoshi, S.; Nishimura, T.; Tsutsumi, H.; Serpone, N. Microwave Discharge Electrodeless Lamps. Part VIII: Continuous On-Site Solar Energy Remediation of Contaminated Water. Chem. Eng. Technol. 2015, 39, 102–107. [CrossRef] 131. Lui, G.Y.; Roser, D.; Corkish, R.; Ashbolt, N.; Jagals, P.; Stuetz, R. Photovoltaic powered ultraviolet and visible light-emitting diodes for sustainable point-of-use disinfection of drinking waters. Sci. Total Environ. 2014, 493, 185–196. [CrossRef] 132. Vassilikogiannakis, G. Singlet Oxygen and Dyes: Synthesis with Visible Light is Where the Future Lies. ChemPhotoChem 2020, 4, 385–387. [CrossRef] 133. Montagnon, T.; Kalaitzakis, D.; Triantafyllakis, M.; Stratakis, M.; Vassilikogiannakis, G. Furans and singlet oxygen—why there is more to come from this powerful partnership. Chem. Commun. 2014, 50, 15480–15498. [CrossRef] 134. Dinda, M.; Maiti, S.; Samanta, S.; Ghosh, P.K. Illustrations of Efficient Solar Driven Organic Reactions. Curr. Org. Synth. 2016, 13, 372–384. [CrossRef] Molecules 2021, 26, 1685 27 of 27

135. Li, P.; Terrett, J.A.; Zbieg, J.R. Visible-Light Photocatalysis as an Enabling Technology for Drug Discovery: A Paradigm Shift for Chemical Reactivity. ACS Med. Chem. Lett. 2020, 11, 2120–2130. [CrossRef] 136. McAtee, R.C.; McClain, E.J.; Stephenson, C.R. Illuminating Photoredox Catalysis. Trends Chem. 2019, 1, 111–125. [CrossRef] 137. Douglas, J.J.; Sevrin, M.J.; Stephenson, C.R.J. Visible Light Photocatalysis: Applications and New Disconnections in the Synthesis of Pharmaceutical Agents. Org. Process. Res. Dev. 2016, 20, 1134–1147. [CrossRef] 138. Nauth, A.M.; Lipp, A.; Lipp, B.; Opatz, T. Sunflow: Sunlight drives fast and green photochemical flow reactions in simple micro- capillary reactors—application to photoredox and H-atom-transfer chemistry. Eur. J. Org. Chem. 2017, 2017, 2099–2103. [CrossRef] 139. Yaseen, M.A.; Mumtaz, S.; Hunter, R.L.; Wall, D.; Robertson, M.J.; Oelgemöller, M. Continuous-flow photochemical transformations of 1,4-naphthoquinones and phthalimides in a concentrating solar trough reactor. Aust. J. Chem. 2020, 73, 1149–1157. [CrossRef] 140. Yaseen, M.A.; Mumtaz, S.; Hunter, R.L.; Wall, D.; Robertson, M.J.; Oelgemöller, M. Corrigendum to: Continuous-flow photochemical transformations of 1,4-naphthoquinones and phthalimides in a concentrating solar trough reactor. Aust. J. Chem. 2020, 73, 1301. [CrossRef] 141. Papakonstantinou, I.; Portnoi, M.; Debije, M.G. The hidden potential of luminescent solar concentrators. Adv. Energy Mater. 2020, 11.[CrossRef] 142. Poliakoff, M.; George, M.W. Manufacturing chemicals with light: Any role in the circular economy? Phil. Trans. R. Soc. A 2020, 378, #20190260. [CrossRef][PubMed] 143. Bochet, C.G. On the Sustainability of Photochemical Reactions. Chim. Int. J. Chem. 2019, 73, 720–723. [CrossRef] 144. Michelin, C.; Hoffmann, N. Photocatalysis applied to organic synthesis—A green chemistry approach. Curr. Opin. Green Sustain. Chem. 2018, 10, 40–45. [CrossRef] 145. Oelgemöller, M.; Oelgemoeller, M. Green Photochemical Processes and Technologies for Research & Development, Scale-up and Chemical Production. J. Chin. Chem. Soc. 2014, 61, 743–748. [CrossRef] 146. Ciana, C.-L.; Bochet, C.G. Clean and Easy Photochemistry. Chim. Int. J. Chem. 2007, 61, 650–654. [CrossRef] 147. Albini, A.; Fagnoni, M.; Mella, M. Environment-friendly organic synthesis. The photochemical approach. Pure Appl. Chem. 2000, 72, 1321–1326. [CrossRef] 148. Ciriminna, R.; Delisi, R.; Xu, Y.-J.; Pagliaro, M. Toward the waste-free synthesis of fine chemicals with visible light. Org. Process Res. Dev. 2016, 20, 403–408. [CrossRef] 149. Oelgemöller, M. The sunny side of chemistry at James Cook Uni. Chemistry in Australia, November 2014; 7.