molecules

Article Heterogenization of Ketone Catalyst for Epoxidation by Low Pressure Plasma Fluorination of Silica Gel Supports

Lucia D’Accolti 1,2,* ID , Nicoletta De Vietro 1, Fiorenza Fanelli 3, Caterina Fusco 2, Angelo Nacci 1,2 ID and Francesco Fracassi 1,3,*

1 Department of Chemistry, University of Bari “Aldo Moro”, Via Orabona 4, 70126 Bari, Italy; [email protected] (N.D.V.); [email protected] (A.N.) 2 ICCOM-CNR, SS Bari, Via Orabona 4, 70126 Bari, Italy; [email protected] 3 NANOTEC-CNR, c/o Department of Chemistry, University of Bari “Aldo Moro”, Via Orabona 4, 70126 Bari, Italy; [email protected] * Correspondence: [email protected] (L.D.); [email protected] (F.F.); Tel.: +39-080-544-2068 (L.D.); +39-080-544-2009 (F.F.)

Received: 1 November 2017; Accepted: 27 November 2017; Published: 30 November 2017

Abstract: Low pressure plasma was used for preparing heterogeneous organocatalysts 2-(A)-(C) suitable for dioxirane-mediated epoxidations. Heterogenization was accomplished by adsorption of the methyl perfluoroheptyl ketone (2) on fluorinated supports (A)-(C) deriving from the treatment of commercial C8-silica gel in low pressure plasma fed with fluorocarbons. Catalyst 2-(C) proved to be the most efficient one, promoting epoxidation of an array of alkenes, including unsaturated fatty esters like methyl oleate (10) and the triglyceride soybean oil (11), with the cheap potassium peroxymonosulfate KHSO5 (caroate) as a green oxidant. Notably, the perfluorinated matrix gives rise to the activation of caroate, generating singlet oxygen. Materials were characterized by infrared Attenuated Total Reflectance spectroscopy (ATR-FTIR), X-ray Photoelectron Spectroscopy (XPS ) and Emission Scanning Electron Microscope (FESEM).

Keywords: dioxiranes; PE-CVD; epoxidation; fluorous chemistry

1. Introduction Fluorous chemistry is a well-known methodology that involves the use of perfluorinated solvents and/or perfluorinated reagents to facilitate catalyst or reaction products recovery [1]. Typically, conventional organic reagents incorporating perfluorinated tags can react in fluorous solvents (e.g., perfluorohexane) and are then extracted into organic phases, or can be supported on fluorous solid matrixes behaving as heterogeneous reactants or catalysts [1,2]. Well-known benefits of this strategy can be found into the increased reaction rates of alkene epoxidation with H2O2 performed into perfluoroalcohols (e.g., 2,2,2-trifluoroethanol or 1,1,1,3,3,3- hexafluoro-2-propanol, HFIP) [2], or with perfluorinated ketones/caroate [3]. However, the use of perfluorinated solvent suffers from the limitation of the prohibitive costs for large-scale applications. To overcome this drawback, Fluorous Silica Gel (FSG) systems, that can be used directly as a catalyst [4], or as solid supports to heterogenize perfluorinated catalytic species by means of fluorous interactions [5,6] are most useful. Representative examples are given by perfluoro-tagged gold nanoparticles immobilized on FSG as suitable catalysts for oxidation of alkenes with H2O2 [7], or immobilized fluorinated flavins useful for promoting the photocatalytic oxidation of benzyl alcohols [8]. In this context, the oxidations mediated by organo-catalyst dioxiranes 1 can find special applications [9,10], alternative to metal catalysis [11,12]. The dioxiranes are very active and selective

Molecules 2017, 22, 2099; doi:10.3390/molecules22122099 www.mdpi.com/journal/molecules Molecules 2017, 22, x 2 of 14 Molecules 2017, 22, 2099 2 of 14 In this context, the oxidations mediated by organo-catalyst dioxiranes 1 can find special applications [9,10], alternative to metal catalysis [11,12]. The dioxiranes are very active and selective oxidants,oxidants, the the most most used used being being dimethyldioxirane dimethyldioxirane (DDO) (DDO) and methyl(trifluoromethyl)dioxirane and methyl(trifluoromethyl)dioxirane (TFDO), which(TFDO), can which be used can either be used as isolated either stoichiometricas isolated stoichiometric reagents [13] orreagents as oxidizing [13] or organocatalysts, as oxidizing generatedorganocatalyst in situs, generated by reacting in situ parent by reacting ketones parent with cheap ketones oxidants with cheap such oxidants as potassium such as caroate potassium [14] (Schemecaroate [114).] (Scheme 1).

SchemeScheme 1.1. Generation of dioxirane inin situsitu fromfrom ketone/caroate.ketone/caroate.

In the last decade, anchoring of the trifluoroacetyl moiety of dioxiranes by grafting onto supports In the last decade, anchoring of the trifluoroacetyl moiety of dioxiranes by grafting onto supports like poly(ethylene) glycol (PEG) or silica gel furnished very active epoxidizing systems of alkenes like poly(ethylene) glycol (PEG) or silica gel furnished very active epoxidizing systems of alkenes [15,16]. [15,16]. In this context, we recently reported the first application of plasma-enhanced chemical vapor In this context, we recently reported the first application of plasma-enhanced chemical vapor deposition deposition (PE-CVD) to the synthesis of supported perfluorinated ketones, by covering Merrifield (PE-CVD) to the synthesis of supported perfluorinated ketones, by covering Merrifield resin beads resin beads with an oxygen-containing fluorocarbon thin film deposited in a hexafluoropropene-O2 with an oxygen-containing fluorocarbon thin film deposited in a hexafluoropropene-O low pressure low pressure non equilibrium plasma [17]. This surface modification approach has to 2be considered non equilibrium plasma [17]. This surface modification approach has to be considered particularly particularly valuable from environmental standpoint since no solvents are required and no valuable from environmental standpoint since no solvents are required and no dangerous wastes are dangerous wastes are produced [18,19]. produced [18,19]. The major advantage of this versatile technique resides in the fact that plasma treatment usually The major advantage of this versatile technique resides in the fact that plasma treatment usually changes the characteristics of the upper layers of the treated material surface without damaging its changes the characteristics of the upper layers of the treated material surface without damaging its bulk properties, thanks to the very low range of penetration. bulk properties, thanks to the very low range of penetration. Interestingly, catalytic activity of supported materials was found to be dependent on the plasma Interestingly, catalytic activity of supported materials was found to be dependent on the plasma deposition conditions, leading either to the selective epoxidation of alkenes (mediated by dioxirane) deposition conditions, leading either to the selective epoxidation of alkenes (mediated by dioxirane) or or preferentially to their oxidative cleavage, depending on whether C3F6-O2 or pure C3F6 were used preferentially to their oxidative cleavage, depending on whether C F -O or pure C F were used to to feed the plasma. 3 6 2 3 6 feed the plasma. Given the success of the original method [17], we decided to extend these findings developing a Given the success of the original method [17], we decided to extend these findings developing new and more simple approach for dioxirane heterogenization, by anchoring its perfluorinated a new and more simple approach for dioxirane heterogenization, by anchoring its perfluorinated parent ketone to new fluorous supports obtained obtained by depositing a perfluoropolymer from parent ketone to new fluorous supports obtained obtained by depositing a perfluoropolymer from perfluoropropene C3F6, analogous to commercial FSG, by simple fluorous interactions. perfluoropropene C F , analogous to commercial FSG, by simple fluorous interactions. The use of low 3pressure6 non equilibrium plasmas assured the fine preparation of the fluorinated The use of low pressure non equilibrium plasmas assured the fine preparation of the fluorinated matrix suitable for a prompt anchoring of dioxirane precursor, which was chosen to be methyl matrix suitable for a prompt anchoring of dioxirane precursor, which was chosen to be methyl perfluoroheptyl ketone C7F15COCH3 (2), a fluorinated ketone previously described as an analogous perfluoroheptyl ketone C7F15COCH3 (2), a fluorinated ketone previously described as an analogous of of trifluoroacetone into the in situ generation of dioxirane [20]. trifluoroacetone into the in situ generation of dioxirane [20]. Anchoring of methyl perfluoroheptyl ketone (2) avoids the drawback of its large scale Anchoring of methyl perfluoroheptyl ketone (2) avoids the drawback of its large scale utilization, utilization, due to its high cost, and overcomes the problem of its low solubility in MeCN/water, due to its high cost, and overcomes the problem of its low solubility in MeCN/water, which imposes which imposes the use of the expensive solvent HFIP [20]. In addition, this approach would permit the use of the expensive solvent HFIP [20]. In addition, this approach would permit the easy the easy preparation and handling of the catalyst, a major efficiency in caroate mediated oxidation preparation and handling of the catalyst, a major efficiency in caroate mediated oxidation without without leaching, and a large scope of olefinic substrates, including fatty acid esters like methyl oleate leaching, and a large scope of olefinic substrates, including fatty acid esters like methyl oleate and and soybean oil. Herein, the synthesis and catalytic activity of this plasma prepared heterogenized soybean oil. Herein, the synthesis and catalytic activity of this plasma prepared heterogenized organocatalyst is disclosed. organocatalyst is disclosed.

2. Results and D Discussioniscussion

Initially, the catalyst supports were prepared by treating commercial C8-silica gel in low pressure Initially, the catalyst supports were prepared by treating commercial C8-silica gel in low pressure plasmas fed with tetrafluoromethane (CF4) or hexafluoropropene (C3F6). The latter reactant typically plasmas fed with tetrafluoromethane (CF4) or hexafluoropropene (C3F6). The latter reactant typically ensures the deposition of thin films with high F-to-C ratios [21]. Tetrafluoromethane, on the other

MoleculesMolecules 20172017,, 2222,, x 2099 33 of of 14 14 ensures the deposition of thin films with high F-to-C ratios [21]. Tetrafluoromethane, on the other Molecules 2017, 22, x 3 of 14 hand,hand, is is generally generally employed employed as as an an etchant etchant gas gas or or for for the the surface surface grafting grafting of of fluorinated fluorinated moieties moieties rather rather than to deposit a fluorocarbon thin films [21]. thanensures to deposit the deposition a fluorocarbon of thin thin films film withs [21 high]. F-to-C ratios [21]. Tetrafluoromethane, on the other Applying these protocols, commercial C8-silica gel was modified leading to the two different hand,Applying is generally these employed protocols, as commercialan etchant gas C or8-silica for the gel surface was mgraftingodified of leading fluorinated to the moieties two different rather perfluorinated supports: (B) bearing grafted perfluorinated chains, and (C) coated with a perfluorinated perfluorinatedthan to deposit supports a fluorocarbon: (B) bearingthin film s grafted[21]. perfluorinated chains, and (C) coated with a thin film (Figure1). For comparison, the commercial perfluorinated silica gel 60 without plasma perfluorinatedApplying thin these film protocols, (Figure commercial 1). For comparison C8-silica gel, the was commercial modified leading perfluorinated to the two silica different gel 60 treatment (A) was also used as support. Afterwards, impregnation of silica gel supports (A)-(C) with withoutperfluorinated plasma treatment supports: ((AB) was bearing also graftedused as perfluorinated support. Afterwards chains, , andimpregnation (C) coated of withsilica a gel methyl perfluoroheptyl ketone (2), afforded the corresponding organocatalyst composites 2-(A), 2-(B) supportsperfluorinated (A)-(C) with thin filmmethyl (Figure perfluoroheptyl 1). For comparison ketone ,( 2the), afforded commercial the corresponding perfluorinated organocatalyst silica gel 60 and 2-(C) to be used as dioxirane precursors in alkene epoxidation. compositeswithout plasma 2-(A), 2 treatment-(B) and 2 (-A(C) ) wasto be also used used as dioxirane as support precursors. Afterwards in ,alkene impregnation epoxidation of silica. gel supports (A)-(C) with methyl perfluoroheptyl ketone (2), afforded the corresponding organocatalyst composites 2-(A), 2-(B) and 2-(C) to be used as dioxirane precursors in alkene epoxidation.

FigureFigure 1. 1. PreparationPreparation of of heterogeneous heterogeneous p perfluorinatederfluorinated organocatalys organocatalysts.ts. Figure 1. Preparation of heterogeneous perfluorinated organocatalysts. Catalytic performances of these materials were evaluated in the model epoxidation reaction of CatalyticCatalytic performances performances of these materials materials were were evaluated evaluated in the in themodel model epoxidation epoxidation reaction reaction of trans-β-methylstyrene (3) with potassium caroate (KHSO5) and compared with those of oftranstrans-β-β-methylstyrene-methylstyrene (3) ( 3) with with potassium potassium caroate caroate (KHSO (KHSO5) 5 and) and compared compared with with those those of of perfluoroketone (2) used alone under homogeneous conditions (Table 1, run 1). perfluoroketoneperfluoroketone (2 ()2 used) used alone alone under under homogeneoushomogeneous conditions (Table (Table 1,1, run 1). 1). Table 1. Epoxidation of alkene 3 with potassium caroate a. TableTable 1. 1.Epoxidation Epoxidation ofof alkene 3 with potassium potassium caroate caroate a. a.

Selectivity (%)c c Run Catalyst Loading b (%) Time (h) Conv. c (%) SelectivitySelectivity (%) (%) c Leaching Run RunCatalystCatalyst LoadingLoading b (%) b (%)TimeTime (h) (h) ConvConv.. c (%)c (%) 3a 3b LeachingLeaching 3a3a 3b3b 1 1 2 2 9.79.7 55 9090 9696 ------2 2 21-(2A-()A ) 2 9.79.7 9.744 57575 9072 9672 –2828 – yesyes 3 3 22-(2B-)( B) 2-(A)9.79.7 9.744 49595 7568 7268 282929 yes yesyes 4 4 23-(2C-)(C ) 2-(B)9.79.7 9.744 49797 9583 6883 291717 yesnonenone 5 5 24-(2C-)(C ) 2-(C)14.614.6 9.722 49595 9786 8386 171414 nonenonenone 6 6 25-(2C-)(C ) 2-(C)9.79.7 14.622 28080 9584 8684 141616 nonenonenone 7 7 26-(2C-)(C ) 2-(C)4.84.8 9.722 29090 8068 8468 163232 nonenonenone 8 8 27-(2C-)(C ) 2-(C)0.00.0 4.822 28080 90- 68 - 329393 nonenonenone 9 d 9 d 28-(2C-)(C ) 2-(C)4.84.8 0.044 29595 809595 – 935 5 nonenonenone d a 9 2-(C) 4.8 4 95 95 5 none a Reaction Reaction conditions conditions: alkene: alkene ( 3()3 )(1 (1 mmol), mmol), heterogeneousheterogeneous catalystcatalyst (100 (100 mg, mg, with with loaded loaded ketone ketone (2 )( 2as) as a ® Reactionindicated conditions), caroate: alkene (4 mmol, (3) (1 mmol), corresponding heterogeneous to 2.86 catalyst mmol/g (100 mg,of Caroat with loaded ® triple ketone salt, (2 iodometry),) as indicated), indicated), caroate (4 mmol, corresponding to 2.86® mmol/g of Caroat triple salt, iodometry),◦ caroate (4 mmol, corresponding tob 2.86 mmol/g of Caroat triple salt, iodometry), CH3CN/aq.c buffer, T = 25 C. CH3CN/aq. buffer, T = 25 °C.b Mol% of dioxirane precursor 2 respect to substrate 3. Evaluatedc by CHb Mol%3CN/aq. of dioxirane buffer, precursorT = 25 °C.2 respect Mol% to of substrate dioxirane3. c Evaluatedprecursorby 2 respect GLC. d Oxidant/substrate to substrate 3. Evaluated ratio of 6:1 wasby used in the presence of NaN3 (30 mg, 0.46 mmol) as an additive.

Molecules 2017, 22, x 4 of 14

d GLC. Oxidant/substrate ratio of 6:1 was used in the presence of NaN3 (30 mg, 0.46 mmol) as an Molecules 2017, 22, 2099 4 of 14 additive. Reaction parameters were calibrated for processing 1 mmol of alkene in the presence of ketone (2) (inReaction the range parameters of 5–15%) and were caroate calibrated (4 mmol) for processing as an oxygen 1 mmol source. of alkeneReaction in medium the presence and ofthe ketone other settings(2) (in the were range chosen of 5–15%) to be quite and caroatesimilar to (4 those mmol) generally as an oxygen adopted source. in the Reaction dioxirane medium epoxidation and thein situ;other in settings particular, were the chosen conditions to be quiteused similarin the case to those of silica generally functionalized adopted inwith the trifluoroacetyl dioxirane epoxidation moiety [16].in situ; Despite in particular, the heterogeneous the conditions conditions, used insupported the case organocatalysts of silica functionalized proved to with be trifluoroacetylhighly active, providingmoiety [16 almo]. Despitest all the higher heterogeneous conversions conditions, in shorter supported reaction times organocatalysts respect to those proved obtained to be highly with methylactive, providingperfluoroheptyl almost ketone all higher (2) alone conversions (Table in1, runs shorter 3–4). reaction As expected times respect for a heterogeneous to those obtained process, with selectivitiesmethyl perfluoroheptyl were slightly ketone lowered, (2) alone due (Table to the1, runs formation 3–4). As of expected benzaldehyde for a heterogeneous (3b) as a byproduct process, resultingselectivities from were the slightly complete lowered, cleavage due to of the the formation double bond. of benzaldehyde In addition, (3b catalysts) as a byproduct 2-(A) and resulting 2-(B) sufferedfrom the from complete the disadvantage cleavage of theof leachi doubleng bond.of the dioxirane In addition, precursor catalysts 2 2from-(A) the and solid2-(B )surface suffered during from thethe reaction disadvantage (Table of 1, leachingruns 2–3). of the dioxirane precursor 2 from the solid surface during the reaction (TableIn1 , contrast, runs 2–3). catalyst 2-(C) showed higher performance, giving the highest conversion and selectivitIn contrast,y without catalyst apparent2-(C )desorption showed higher of methyl performance, perfluoroheptyl giving the ketone highest ( conversion2) (no leaching and selectivity, Table 1, runwithout 4). Therefore, apparent 2 desorption-(C) was used of methylin the successive perfluoroheptyl experiments ketone, where (2) (no the leaching, influence Table of the1, catalyst run 4). loadingTherefore, was2 -(investigatedC) was used by in using the successive variable amounts experiments, of (2) where adsorbed the influenceon support of (Table the catalyst 1, runs loading 5–7). wasFrom investigated the screening by using clearly variable emerged amounts that of the (2) increase adsorbed of on that support adsorbed (Table precursor1, runs 5–7). up to 15% ca. did notFrom result the in screening an increment clearly of emerged conversion that and the increaseselectivity of, thatbut solely adsorbed in a precursorreduced reaction up to 15% time ca. (Tabledid not 1, resultrun 5) in. In an contrast, increment the ofdecrease conversion of (2) and down selectivity, to 5% ca. but led solelyto the information a reduced of reactionmuch higher time amounts(Table1, runof the 5). oxidative In contrast, cleavage the decrease compound of ( 2 )benzaldehyde down to 5% ca. (3b led) (Table to the 1, formation run 7). In of line much with higher this trendamounts, when of thereaction oxidative was performed cleavage compound in the presence benzaldehyde of the sole ( 3bperfluorinated) (Table1, run support 7). In line (C), with (3b) thiswas thetrend, sole when product reaction observed was performed(Table 1, run in 8). the presence of the sole perfluorinated support (C), (3b) was the soleThe productrationale observed for this switching (Table1, run of selectivity 8). is the probable activating effect exerted by fluorous silica The gel rationale(prepared for by this plasma switching processing of selectivity) towards is the probablethe oxygen activating source effectcaroate exerted. The byactivating fluorous mechanismsilica gel (prepared, not yet by completely plasma processing) understood, towards assume the oxygens that source perfluoroalkyl caroate. The chains activating on the mechanism, support surfacenot yet completelycould share understood, the fluorine assumes atoms affording that perfluoroalkyl a multiple chainsweak hydrogen on the support-bond surface network could [4], sharethus establishingthe fluorine atomsa strong affording interaction a multiple with caroate weak hydrogen-bondthat should lead network to a more [4], facile thus establishingO–O peroxide a strongbond cleavageinteraction (Figure with caroate2). that should lead to a more facile O–O peroxide bond cleavage (Figure2).

FigureFigure 2 2.. ProposedProposed mechanism mechanism for for caroate caroate activation activation by by fluorinated fluorinated silica silica gel gel support support..

The activated KHSO5 should immediately react with adsorbed ketone (2), leading to epoxidation The activated KHSO5 should immediately react with adsorbed ketone (2), leading to epoxidation viavia dioxirane dioxirane (1 ()1 (scheme) (Scheme 1).1 Otherwise,). Otherwise, in the in absence the absence of the methyl of the perfluoroheptyl methyl perfluoroheptyl ketone, it ketone, should 1 undergoit should the undergo self-decomposition the self-decomposition reaction according reaction to according Equation to (1 Equation), thus releasing (1), thus single releasing oxygen single O2 which is 1probably responsible for the oxidative cleavage of the double bond to give benzaldehyde oxygen O2 which is probably responsible for the oxidative cleavage of the double bond to give [benzaldehyde3,22,23]. [3,22,23]. − − 2− 1 HSO5 + SO5 → HSO4 + SO4 + O2, (1) , (1) This latter hypothesis was confirmed by the reaction outcome of epoxidation carried out in This latter hypothesis was confirmed by the reaction outcome of epoxidation carried out in the the presence of the singlet oxygen quencher NaN3, for which selectivity turned to be completely in presencefavor of of the the epoxide singlet productoxygen quencher (3a) (Table NaN1, run3, for 9) which [ 24,25 selectivity]. Note toturned mention, to be thecompletely activating in favor effect of the perfluorinated matrix allowed to achieve a double advantage: (i) a higher reactivity of heterogeneous catalyst 2-(C) respect to methyl perfluoroheptyl ketone (2) and (ii) a major flexibility Molecules 2017, 22, x 5 of 14

of the epoxide product (3a) (Table 1, run 9) [24,25]. Note to mention, the activating effect of the Molecules 2017, 22, 2099 5 of 14 perfluorinated matrix allowed to achieve a double advantage: (i) a higher reactivity of heterogeneous catalyst 2-(C) respect to methyl perfluoroheptyl ketone (2) and (ii) a major flexibility of the protocol, ofenabling the protocol, the possibility enabling of the performing possibility two of performing different kind twos of different olefin transformations kinds of olefin transformations (epoxidation or (epoxidationdouble bond orcleavage double) bondby means cleavage) of the by simpl meanse choice of the of simple starting choice reagents of starting. reagents. After thethe optimizationoptimization experiments,experiments, the novelnovel heterogeneousheterogeneous perfluorinatedperfluorinated silica gelgel plasmaplasma coatedcoated catalystcatalyst 2-(-(CC)) was fully characterized by ATR-FTIR,ATR-FTIR, XPSXPS andand SEM techniques.techniques. Infrared spectra inin FigureFigure3 3 clearly clearly confirmedconfirmed the the adsorption adsorption ofof ((2) on the surface of plasma treated support (C) evidenced byby thethe carbonylcarbonyl band,band,centered centered at at 1780 1780 cm cm−−11 ν (C=O)(C=O),, which which appeared appeared into the spectrum ofof coatedcoated silicasilica gelgel afterafter impregnationimpregnation [[2266,27,27].].

Figure 3. ATR-FTIR spectra of C Silica gel (A) pristine, (B) after plasma coating and (C) after plasma Figure 3. ATR-FTIR spectra of C88 Silica gel (A) pristine, (B) after plasma coating and (C) after plasma coating and ketone (2) impregnation. coating and ketone (2) impregnation.

The XPS atomic composition of the supports surface (Table2 2)) clearly clearly evidenced evidenced the the high high degree degree of fluorination obtainedobtained inin thethe casecase ofof the coated support (C),), as indicated byby thethe high F concentration (Table(Table2 ,2, run run 4). 4). ThisThis shouldshould explainexplain thethe superiorsuperior catalyticcatalytic performanceperformance ofof thethe impregnatedimpregnated catalystcatalyst 2-(-(CC)) together together with with its its better better capacity capacity in retaining in retaining adsorbed adsorbed methyl methyl perfluoroheptyl perfluoroheptyl ketone ( ketone2) without (2) leaching.without leaching. It is worth It is mentioning worth mentioning that the XPSthat datathe XPS also data evidenced also evidenced that during that impregnation, during impregnation, a partial detachmenta partial detachment of the fluorinated of the fluorinated coating occurs, coating as demonstrated occurs, as demonstrated by the increase by of atomicthe increase concentration of atomic of theconcentration underlying of silicon the underlying from 1% to 7%silicon (Table from2, runs 1% 4–5,to 7% see (Table also Suppo 2, runs Info 4– Table5, see S1,also Figures Suppo S1 Info and Table S2). S1 and Figure S1, S2). Table 2. XPS Atomic composition of supports and catalyst 2-(C) a. Table 2. XPS Atomic composition of supports and catalyst 2-(C) a. Run Support C% O% Si% F% Run Support C% O% Si% F% 1 C8-silica gel 27 ± 2 49 ± 1 24 ± 1 / 1 C8-silica gel 27 ± 2 49 ± 1 24 ± 1 / 2 Perfluorin. silica gel 60 (A) 17 ± 3 42 ± 2 24 ± 2 17 ± 3 32 Perfluorin.Perfluorin. silica silica gel gel grafted 60 (A (B)) 1917 ±± 32 42 35 ± ±2 324 22± 2± 317 ± 24 3 ± 3 43 Perfluorin.Perfluorin. silicasilica gelgel coatedgrafted (C)(B) 4119 ±± 23 352 ±± 3 122 ± 1 3± 124 ± 56 3 ± 3 54 Perfluorin. Catalyst silica 2-(C)gel coated (C) 3241 ±± 3 226 ± 1± 31 ± 7 1± 256 ± 35 3 ± 3 5 Catalysta Maximum 2-(C) relative standard32 deviation ± 3 26 3% ± ca. 3 7 ± 2 35 ± 3 a Maximum relative standard deviation 3% ca. Finally, thethe scanningscanning electronelectron microscopymicroscopy (SEM)(SEM) imagesimages reportedreported inin FigureFigure4 4 show show that that the the silica silica surfacesurface waswas coatedcoated byby thethe perfluoroperfluoro polymerspolymers (Figure(Figure4 4B)B). . The stability of catalyst 2-(C) was studied in a series of recycling experiments (Figure5). Being the ketone (2) solely adsorbed on support surface by means of fluorous interactions, recycling procedures Molecules 2017, 22, x 6 of 14

Molecules 2017, 22, 2099 6 of 14 were accomplished in a careful manner to avoid desorption. After each run, the solid residue was decanted and supernatant aqueous buffer was removed with care washing with little aliquots of fresh solvent. Then, fresh buffer and reagents were added to the solid for a new run (see Section 3.4). AsMolecules shown 2017 in, 22 Figure, 2099 5, catalytic performances for the first five runs remained above 80% of conversion.6 of 13

(A)

(B) Figure 4. (A) Pristine silica powder; (B) silica powder with perfluoro plasma polymer.

The stability of catalyst 2-(C) was studied in a series of recycling experiments (Figure 5). Being the ketone (2) solely adsorbed on support surface by means of fluorous interactions, recycling procedures were accomplished in a careful manner to avoid desorption. After each run, the solid residue was decanted and supernatant aqueous buffer was removed with care washing with little aliquots of fresh solvent. Then, fresh buffer and reagents were added to the solid for a new run (see Section 3.4). As shown in Figure 5, catalytic performances for the first five runs remained above 80% Figure 4. (A) Pristine silica powder; (B) silica powder with perfluoro plasma polymer. of conversionFigure. 4. (A) Pristine silica powder; (B) silica powder with perfluoro plasma polymer.

100 100

80 80

60 (%) 60

40 40

Conversion (%) Conversion Conversion 20 20

0 0 12345 1 2 3 4 5 run run

Figure 5. Recycling of 2-(C) organocatalysts in the epoxidation of alkene 3 with caroate. Figure 5. Recycling of 2-(C) organocatalysts in the epoxidation of alkene 3 with caroate. Figure 5. Recycling of 2-(C) organocatalysts in the epoxidation of alkene 3 with caroate. The evident reduction of catalytic activity was due to both the partial desorption of methyl perfluoroheptylThe evident ketone reduction (detected of catalytic by GLC activity analyses was due of mother to both liquor) the partial and desorption the accumulation of methyl of perfluoroheptylpotassium sulfate ketone byby-product.-product. (detected by GLC analyses of mother liquor) and the accumulation of potassium sulfateHoweverHowever, by-product., to evaluate the whole catalyst robustnessrobustness,, epoxidationepoxidation ofof transtrans-β--methylmethyl styrene (3) was carriedHowever, outout towith with evaluate excess excess theamounts amounts whole of catalyst of alkene alkene robustness,and and monitored monitored epoxidation until un catalysttil catalyst of trans deactivation -deactivationβ-methyl (see styrene (seeSection S (3ection) 3.5). was 3.5)carriedUnder. Under these out withthese conditions excess conditions amountsa total a tota turnover ofl alkeneturnover number and number monitored (TON) (TON of until 206) of catalyst206ca. was ca. was observed, deactivation observed indicating (see, indicating Section that that 3.5the). theUndercatalyst catalyst these is stable is conditions stable and and much a totalmuch more turnover more active active number respect respect (TON) to to its its of homogeneous 206homogeneous ca. was observed, counterpartcounterpart indicating [17][17] (see(see that also the Tablecatalyst 11,, isrun stable 11).). and much more active respect to its homogeneous counterpart [17] (see also Table1, run 1).The loss of catalytic activity of 2-(C) was explained with the partial removal of plasma fluoropolymer from the surface of silica support as a consequence of reaction conditions; most probably due to the mechanical damage of stirring (Figure 6A). This assumption was verified re-submitting deactivated catalyst to C3F6–O2 plasma treatment that restored the perfluoropolymer coating and consequently, after re-impregnation of methyl perfluoroheptyl ketone (2), the catalyst performance, as demonstrated by the high conversion (95%) and selectivity (90%) displayed by the regenerated 2-(C) in the epoxidation of (3).

Molecules 2017, 22, 2099 7 of 14

The loss of catalytic activity of 2-(C) was explained with the partial removal of plasma fluoropolymer from the surface of silica support as a consequence of reaction conditions; most probably due to the mechanical damage of stirring (Figure6A). This assumption was verified re-submitting deactivated catalyst to C3F6–O2 plasma treatment that restored the perfluoropolymer coating and consequently, after re-impregnation of methyl perfluoroheptyl ketone (2), the catalyst performance, as demonstrated by the high conversion (95%) and selectivity (90%) displayed by the regenerated 2-(C) Moleculesin the epoxidation 2017, 22, 2099 of (3). 7 of 13

(A)

(B)

Figure 6. ((AA)) Exhausted Exhausted catalyst catalyst 2-((C);); ( (B)) regenerated regenerated catalyst 2-(C).

The substrate scope of 2-(-(C) was was studied studied with with an an array array of of olefin olefin substrates substrates (Table (Table3 3).). EpoxidationEpoxidation of cyclohexene cyclohexene (4 ()4 took) took place place with with 98% 98% of conversion of conversion after after2 h and 2 hthe and cyclohexene the cyclohexene oxide (4a) oxide product (4a) wasproduct obtained was with obtained excellent with selectivity excellent (Table selectivity 3, run (Table 1). Similarly,3, run 1). the Similarly,allylic alcohol the cyclohex-2-enol allylic alcohol (cyclohex-2-enol5) reached full conversion, (5) reached afford full conversion,ing the corresponding affording the epoxide corresponding (5a) together epoxide with a (5a 15%) together of cyclohex- with 2-enonea 15% of (5b cyclohex-2-enone) [28] (Table 3, run (5b 2).)[ In28 ]contrast, (Table3 ,1-octene run 2). ( In6) contrast,afforded only 1-octene a 15% ( 6 of) affordedconversion only (Table a 15% 3, runof conversion 3), while no (Table epoxidation3, run 3), occurred while no with epoxidation the more occurred deactivated with substrate the more trans deactivated-cinnamonitrile substrate (7), whichtrans-cinnamonitrile rather partially (7 underwent), which rather the double partially bond underwent cleavage the affording double bondbenzoic cleavage acid (Table affording 3, run benzoic 4). acidExtending (Table3, run the 4). substrate scope to the selective oxidation of key organic compounds, the protocol was alsoExtending applied the to substratea representative scope to substrate the selective of each oxidation of the following of key organic class of compounds, compounds: the an protocol alkyne (3-hexynewas also applied (8)), an to alcohol a representative (3-hexanol substrate (9)), an unsaturated of each of the natural following monoester class of (methyl compounds: oleate an(10 alkyne)) and an(3-hexyne unsaturated (8)), antriglyceride alcohol (3-hexanol (soybean (oil9)), (11 an)). unsaturated natural monoester (methyl oleate (10)) and an unsaturatedOxidation of triglyceride 3-hexyne ( (soybean8) afforded oil smoothly (11)). the expected 3,4 hexanedione (8a) (Table 1, run 5), a veryOxidation interesting of 3-hexyne result if (8considering) afforded smoothly that 1,2-diones the expected are valuable 3,4 hexanedione building ( 8ablocks) (Table in1 ,organic run 5), synthesis,a very interesting especially result as ifprecursors considering of thatseveral 1,2-diones heterocyclic are valuable compounds building [29–31]. blocks inThis organic representative synthesis, exampleespecially shows as precursors how the of severalnovel catalyst heterocyclic is more compounds reactive [29 than–31 ].those This representativereported previously example [16,20], shows includinghow the novel also the catalyst methyl is perfluoroheptyl more reactive than ketone those 2 used reported under previously homogeneous [16,20 conditions], including [20]. also the methylIn contrast, perfluoroheptyl 3-hexanol ketone (9) afforded2 used under only trace homogeneous amounts conditionsof expected [ 20ketone]. (9a), even after using excessIn amounts contrast, of 3-hexanol caroate (and9) afforded prolonged only reaction trace amounts times (48 of h). expected This indicates ketone ( 9athat), eventhis protocol after using is highlyexcess selective amounts towards of caroate epoxidation and prolonged and can reaction tolerate timesother (48functional h). This groups indicates susceptible that this to protocoloxidation. is highlyFinally, selective epoxidation towards epoxidation of fatty acid and esters can (10 tolerate)and (11 other) (Table functional 3, runs groups 7–8) [32] susceptible further validates to oxidation. the efficacyFinally, of this epoxidation method that of can fatty be acid used esters even (in10 the)and production (11) (Table of3, epoxidized runs 7–8) [ 32soybean] further oil validates (ESO) (11a the), aefficacy bulk chemical of this method of great that practical can be used importance even in thelargely production used as of plasticizer, epoxidized lubricant, soybean oil cross-linking (ESO) (11a), agent, stabilizer etc. [33].

Table 3. Substrate scope in the alkene epoxidation promoted by catalyst 2-(C) a.

catalyst 2-(C)/KHSO5 substrate oxidation products CH3CN / aqueous phosphate buffer 4-11 pH 7-8, 25 °C Oxd./Sub. b Time Conv. c Run Substrate Products (Yield %) d Ratio (h) (%)

1 4 2 98

Molecules 2017, 22, 2099 8 of 14

Molecules 2017, 22, x 8 of 14 a bulk chemicalMolecules 2017 of, 2 great2, x practical importance largely used as plasticizer, lubricant, cross-linking8 of 14 agent, Molecules 2017, 22, x 8 of 14 MoleculesMolecules 2017, 2 22017, x , 22, x 8 of 14 8 of 14 stabilizerMolecules etc.Molecules [ 201733]., 2 201722017, x , , 2222Table,, xx 3. Substrate scope in the alkene epoxidation promoted by catalyst 2-(C) a. 8 of 14 88 of 1414 MoleculesMolecules 2017,, 2 20172Table,, xx , 22 ,3 x. Substrate scope in the alkene epoxidation promoted by catalyst 2-(C) a. 8 of 14 8 of 14 MoleculesMolecules 20172017,,, 22 220172,,, xxx , , 22,, xx 88 ofof 1414 8 of 14 Table 3. Substrate scope in the alkene epoxidation promoted by catalysta 2-(C) a. Table 3. Substrate scope in the alkene epoxidation promoted by catalyst 2-(C) .a a Table Table3. Substrate 3. Substrate scope scopein the inalkene the alkene epoxidation epoxidation promoted promoted by catalyst by catalyst 2-(C) 2. -(C) a.a Table 3. SubstrateTable 3. scopeSubstrate in scopethe alkene in the alkeneepoxidation epoxidation promoted promoted by by catalyst catalyst a 22--(C)) a . . TableTable 3. Substrate Table3. Substrate 3. scopeSubstrate scope in scope thein the alkene inalkene the alkeneepoxidation epoxidation epoxidation promoted promoted promoted by catalyst by by catalyst catalyst 2-(C) 2 2.-( -(CC) a. . Table Table3.. Substrate 3. Substrate scope scopeinin thethe inalkene the alkene epoxidation epoxidation promoted promoted by catalyst by catalyst 2--((C)) aa2.. -(C) a. Table Table3. Substrate 3. Substrate scope scopein the inalkene the alkene epoxidation epoxidation promoted promoted by catalyst by catalyst 2-(C) 2. -(C) .

Oxd./Sub. b Time Conv. c Run Substrate Oxd./Sub. b Time Conv. c Products (Yield %) d d Run Substrate Oxd./Sub.Rbatio b Time(h) Conv.c (%) c Products (Yield %) Oxd./Sub.Ratio b Time(h)b Conv.(%) c c d d Run Run SubstrateSubstrate Oxd./Sub.Oxd./Sub. Time b TimeConv. Conv. c ProductsProducts (Yield (Yield%) %) Oxd./Sub.b b Time Conv.c c d d Run Run SubstrateSubstrate Oxd./Sub.RatioOxd./Sub. Ratio Time(h) b Time(h)Conv.(%) Conv. (%) c ProductsProducts(Yield (Yield %) %) d Run Substrate Oxd./Sub.bbb b b Time Conv.cc c c c Products (Yieldd %) d Run Run SubstrateSubstrate Oxd./Sub.Oxd./Sub.Oxd./Sub.RatioOxd./Sub. R atio Time (h)Time Time (h)Conv. Conv.(%) Conv. Conv. (%) ProductsProducts (Yield (Yield %) d %) d Run RunRun Run1 SubstrateSubstrate Substrate Ratio Ratio4 (h) (h)(h)2 (%) (%)(%)98 ProductsProductsProductsProducts (Yield(Yield (Yield(Yield (Yield%)%) dd %)%)%) %) dd d 1 RRatioRatioatio4atio Ratio (h)(h)(h)2 (h) (h)(h) (%)(%)98(%) (%)(%) (%) 1 4 2 98 1 4 2 98 1 1 4 4 2 2 98 98 1 1 4 4 2 2 98 98 1 11 1 44 44 22 22 9898 9898

1 2 4 4 2 1 98 98 2 4 1 98

2 2 4 4 1 1 98 98 2 2 4 4 1 1 98 98 2 2 4 4 1 1 98 98 2 2 2 4 44 1 11 98 9898 2 32 4 84 1 121 98 1598 3 8 12 15

2 3 3 48 8 121 12 15 98 15 3 3 8 8 12 12 15 15 3 3 8 8 12 12 15 15 3 3 3 8 88 12 1212 15 1515 33 43 88 48 1212 4812 1515 1615 4 4 48 16

4 4 4 4 48 48 16 16 3 4 4 8 4 4 4812 48 1615 16 4 4 4 4 44 48 4848 16 1616 4 4 4 4 48 48 16 16 44 4 44 4 4848 48 1616 16 5 8 4 98 5 8 4 98 5 5 8 8 4 4 98 98 4 5 5 4 8 8 484 4 9816 98 5 5 5 8 88 4 44 98 9898 5 5 8 8 4 4 98 98 55 5 88 8 44 4 9898 98

6 12 48 2 6 12 48 2 6 6 12 12 48 48 2 2

5 6 6 6 812 12 12 484 4848 2 982 2 6 6 12 12 48 48 2 2 6 6 12 12 48 48 2 2 6 6 12 12 48 48 2 2

7 4 4 95 7 7 4 44 495 95 7 7 4 4 4 4 95 95 7 7 4 4 4 4 95 95 6 7 7 124 4 484 4 95 2 95 7 7 4 4 4 4 95 95 77 7 44 4 44 4 9955 95

8 8 1010 99 9595 8 10 9 95 8 8 10 10 9 9 95 95 8 8 10 10 9 9 95 95 7 8 8 410 10 94 9 959 5 95 88 8 1010 10 99 9 9595 95 a Reaction conditionsa Reaction : substrate conditions (1: substrate mmol), 100(1 mmol), mg of 100 catalyst mg of 2catalyst-(C) (bearing 2-(C) (bearing 9.7 × 10 9.7− 2× mmol10−2 mmol of 2 ofper 2 per 100 100 mg), a × −2 Reaction conditions: substrate® (1 mmol), ®100 mg of catalyst 2-(C) (bearing 9.7 ◦ 10b mmol of 2 perb 100 amg), caroate as indicated (Caroat triple salt, iodometry), CH3CN/aq. buffer, T −=2 25 °C. KHSO5 to caroatea as Reaction indicated Reaction conditions (Caroat conditions : substratetriple : substrate salt,(1 mmol),® iodometry), (1 mmol), 100 mg 100 of CH catalystmg3CN/aq. of catalyst 2-(C buffer,) ( bearing2-(C) T (bearing = 9.7 25 ×C. 10 9.7−2 KHSO mmol× 10 mmolof5b to 2 per substrate of 100 2 per 100 molar mga ), caroatea as indicated (Caroat triple salt, iodometry), CH3CN/aq. buffer, T =− 225 °C.−2 KHSO5 to c Reactiona Reaction conditions conditionsd: substrate: substratec (1 mmol), (1e mmol), 100 mg 100 ofd mgcatalyst of catalyst 2-(C) (2bearing-e(C) (bearing 9.7 × 10 9.7 ×mmol 10 −2 mmolof 2 per of 1002 per 100 ratio. Evaluateda substratea Reaction by GLC. mconditionsolarIsolated ratio.: substrate Evaluated yields.® (1 mmol),Mixture® by GLC. 100 of syn/antimgIsolated of catalyst yields. stereoisomers 2 -( CMix) (bearingture in of× a 60:40s9.7yn/anti−2 × 10 ratio− 2stereoisomers bmmol ca.,asb of determined2 per in 100 a mg Reaction), caroatemga Reaction), conditions caroate as indicated conditions as: substrate indicatedc :( Caroatsubstrate (1 ( Caroatmmol), triple (1 mmol), 100salt,tripled mg iodometry), 100 salt, of mgcatalyst iodometry), of catalyst CH 2-e(C3CN/aq.) (CH2bearing-(C3CN/aq.) buffer,(bearing 9.7 buffer,× T 10 9.7= 25 ×mmol T10°C. =− 2 25 mmol ofKHSO °C. 2 per of KHSO5 1002to per 5100 to substrateaa Reactiona Reaction mconditionsolar conditionsratio.: substrate Evaluated: substrate (1 mmol), by® (1 GLC. mmol),® 100 mgIsolated 100 of mgcatalyst yields. of catalyst 2 -( CMix) (2bearingt-ure(C) (ofbearing s9.7yn/anti × 10 9.7−− 22stereoisomers ×mmol 10−2 mmolofb 2 per ofb in 1002 aper 100 mg Reaction),f caroatemg Reaction ),conditions caroate as conditionsindicated :as: substratesubstrate indicated: (substrateCaroat (1(1 (mmol),mmol),Caroat® triple (1 mmol), ® ®100100 triplesalt, mgmg iodometry),100 salt, fofof mgcatalystcatalyst iodometry), of catalyst g22CH--((C3))CN/aq. ( (CH2bearingbearing-(C3CN/aq.) (buffer,bearing h 9.79.7 ×buffer, 10T10 9.7 = 25mmol× T10 °C. = 25 mmolofofb KHSO °C.22 per bofb KHSO1005 1002 to perg 5100 to by GC-MS.mg), Acaroate60:40mg 15%), caroateratio of as cyclohexanone indicatedca. as, asc indicated determined (Caroatc was(Caroat® tripleby detected GC triplesalt,-dMS byiodometry),. salt, dA GC-MS. 15% iodometry), of cyclohexanoneCHeGLC3CN/aq. CH yield.e 3CN/aq. buffer, wasEvaluated buffer, Tdetected = 25 T °C. by= by 25b NMR KHSOGC°C. -MS (Figure KHSO5 to. GLC5 S3to in substratemg), caroatesubstratemg m), olarcaroate as m ratio.indicatedolar as ratio.indicatedEvaluated (Caroat Evaluated ( Caroatby®® triple GLC. by® triplesalt,f GLC. Isolated iodometry), salt, Isolated iodometry),yields. CHyields. Mix3CN/aq. CHt ure Mix3CN/aq. of buffer,t uresyn/anti ofbuffer, Ts yn/anti= stereoisomers 25 T °C. = stereoisomers25 bb KHSO°C.g bin KHSO5 ato in5 toa 60:40mg),), caroateratiocaroatemg), ca.caroate as, as indicated determined as indicatedc ((Caroatc by (GCCaroat® tripletriple-MS®. salt,triplesalt, Ad 15% iodometry),iodometry), salt,d of iodometry),cyclohexanone CHCHe 3CN/aq. CHe was3CN/aq. buffer, detected buffer, T = by25 T GC°C. = 25-bMS KHSO °C.. bGLC KHSO5 toto 5 to Supportingsubstratemg Information)), caroatesubstratemg ),m caroateolarh as mindicatedratio.olar as indicatedratio.Evaluated (Caroat c Evaluated (Caroat by triple GLC. by triplesalt, GLC. Isolated iodometry), salt, d Isolated iodometry), yields. CHyields. Mix33CN/aq. CH turee Mix3CN/aq. ofbuffer,ture syn/anti ofbuffer, T s yn/anti= stereoisomers25 T °C. = stereoisomers25 KHSO°C. inKHSO55 to a in5 toa yield.substrate Evaluated molar cratio. by NMR c Evaluated (Figure by S3f dGLC.in supportingf d Isolated Information) yields.e e Mix ture of syn/anti stereoisomersg g in a 60:40substrate ratio60:40hsubstrate mca. ratioolar, as m ratio.ca.determinedolar, as ratio. determinedEvaluated c byEvaluated GC by by-MS GLC. GC .by -A MSGLC. 15%Isolated. A dof Isolated15% cyclohexanone yields. of cyclohexanone yields. Mix teure wasMix of tdetectedure swasyn/anti of detectedsyn/anti bystereoisomers GC by-stereoisomersMS GC. -GLCMS in. a GLC in a 8 yield. substrate Evaluated m byolar NMR ccratio. (Figure c Evaluated S3 in10 supportingby f dGLC.d f d Isolated Information)9 yields.ee 95 e Mixture of syn/anti stereoisomersg g in a substrate60:40substrate ratio60:40substrate m ca.olarratioolar, as m ratio. ratio.ca. olardetermined, as ratio. Evaluateddetermined Evaluated by GCby by -GLC.MS GC by. - AGLC.MS IsolatedIsolated 15%. f A ofIsolated Isolated15% cyclohexanoneyields.yields. of cyclohexanoneyields.yields. Mixt t ure Mixwas oftt ure detecteds yn/antiyn/antiwas of detectedsyn/anti stereoisomersstereoisomersby GC stereoisomersby-MS GC. - MSGLC inin .aa g GLC in a h60:40 hratio ca., as determined by GCff -MS. f A 15% of cyclohexanone was detected by GCg-MS. g GLC yield.60:40 yield.ratio60:40 Evaluated ca.ratio Evaluated, as ca. bydetermined, NMRas bydetermined (FigureNMR by (FigureGC S3 by -inMS GCsupporting S3. - AMSin 15%supporting. f A of Information)15% cyclohexanone of Information) cyclohexanone was detected was detected by GC by-MS GC. - MSGLC. g GLC 60:40h ratioh ca., as determined by GCfff-MS. f A 15% of cyclohexanone was detected by GCgg-MS. g GLC 60:40yield.60:40 ratioratioyield.60:40 Evaluated ca. ca. ratioh Evaluated,, asas ca. determineddeterminedby,, asasNMR determineddeterminedby NMR(Figure byby GC(Figure S3 byby--MS in GC . supporting. S3 --AMS in 15%15% supporting.. A ofof 15% Information) cyclohexanonecyclohexanone of Information) cyclohexanone waswas detecteddetected was detected byby GCGC by--MS GC.. - -GLCMS.. GLC 3. Experimentalhyield. h Evaluated Section by NMR (Figure S3 in supporting Information) 3. Experimentalyield. yield. Section Evaluated h Evaluated by NMR by NMR(Figure (Figure S3 in supportingS3 in supporting Information) Information) 3. Experimentalyield.yield. yield.hh Evaluated hS Evaluatedection by NMR by (FigureNMR(Figure (Figure(Figure S3S3 inin supporting supporting S3S3 inin supportingsupporting Information)Information) Information)Information) 3. Experimental3. Experimental Section Section 3. Experimental3. Experimental Section Section 3.1. Generala Reaction3. Experimental Information3.13. conditionsExperimental. General SInformation: ectionsubstrate Section (1 mmol), 100 mg of catalyst 2-(C) (bearing 9.7 × 10−2 mmol of 2 per 100 3.13.3. Experimental. General3. Experimental Information Sectionection S ection 3.1. General3.1. General Information Information ® b mg3.1), caroate. General3.1. GeneralThe as Information indicatedgas Information chromatography (Caroat triple (GLC) salt, analyses iodometry), were performed CH3CN/aq. using buffer, a GC T- 2010= 25 Plus°C. ( SHIMADZUKHSO5 to The3.1 gas. GeneralThe3.1 chromatography .. gas General Information chromatography Information (GLC) (GLC) analyses analyses werewere performed performed using using a GC a- GC-20102010 Plus ( PlusSHIMADZU (SHIMADZU substrate3.13.1.. GeneralTheDuisburg3.1 m ..gas olarGeneralThe Information chromatography ratio.gas, Germany Information chromatography c Evaluated ) gas (Gchromatograph LbyC) (G GLC.analysesLC) danalyses Isolated wereequipped performedwere yields. withperformed e a Mix usingflameture using aionization GCof s-a2010yn/anti GC -Plusdetector2010 stereoisomers (PlusSHIMADZU (FID) (SHIMADZU and in a ZBa - DuisburgThe ,gas TheGermany chromatography gas chromatography) gas chromatograph (GLC) (G analysesLC) equipped analyses were with wereperformed a performedflame usingionization using a GC detector-a2010 GC -Plus2010 (FID) (PlusSHIMADZU and (SHIMADZU a ZB- Duisburg, Germany)The1 column gasThe chromatography gas(100% gas chromatography chromatograph dimethylpolysiloxane (GLC) (G analysesL equippedC)f analyses stat wereionary with wereperformed phase, aperformed flame 30 usingm ionization× 0.2 using a µGCm - aID)2010 GC detector utilizing -Plus2010 (PlusSHIMADZU n (FID)-butanol (SHIMADZU andg as athe ZB-1 60:40Duisburg ratioTheDuisburg ca. , gasGermanyThe, aschromatography ,gas Germanydetermined chromatography) gas chromatograph) gas by chromatograph(G GCLC) -(GMS analysesL. C) equipped Aanalyses 15% equippedwere of with wereperformedcyclohexanone awith flameperformed a flameusingionization usingwas ionizationa GC detecteddetector -a2010 GC detector -Plus2010 (FID)by ( PlusSHIMADZUGC (FID)and- MS(SHIMADZU a and.ZB GLC- a ZB - 1Duisburg columnTheDuisburg gas(100%, TheGermany chromatography gas, dimethylpolysiloxaneGermany chromatography) gas chromatograph) gas (GchromatographLC) (G analyses statLC) equippedionary analyses wereequipped phase, with wereperformed 30 awith performed flamem × a 0.2 usingflame ionization µm using µ aionizationID) GC utilizing --detectora20102010 GC - detectorPlusPlus2010 n (FID)-butanol ((PlusSHIMADZU (FID)and (SHIMADZU as a and ZBthe- a ZB- column1Duisburg (100%columnh 1internalDuisburg column dimethylpolysiloxane(100%, Germany standard (100%, , dimethylpolysiloxane Germany) dimethylpolysiloxanegas for chromatograph)) gas calibration. chromatograph stationary stat The equippedionary stat GC equippedionary phase,- phase,MS with (gas phase, 30 30 awith chromatographymflame m ×30 a×0.2 flamem ionization0.2 µ ×m 0.2 ID)ionizationm µm utilizing ID)- detectormassID) utilizing utilizing detector s pectrometry)n (FID)-butanol n n-butanol(FID)-andbutanol as a and theZB analyses as - a ZB asthe-- the yield.internalDuisburg1 column Evaluated1Duisburg column standard ,,(100% Germany ,,(100% byGermanydimethylpolysiloxane for NMR)) gasdimethylpolysiloxane calibration. chromatograph)(Figure gas chromatograph S3 The in stat GC supporting equippedequipped-ionaryMS stat equipped (gasionary phase, with Information) chromatography phase, 30 awith flame m 30×a 0.2flame mionization µ× m0.2- massionization ID) µm utilizing detector sID)pectrometry) utilizing detector n(FID)-butanol n (FID)and - analysesbutanol asa and ZB the -- asa ZBthe- internalinternal1 standard columninternalwere1 column standard (100% run for standard on (100% calibration. dimethylpolysiloxane a for GCMS calibration.dimethylpolysiloxane for QP2010 calibration. The The GC-MSSE ( SHIMADZU GCstat The-ionaryMS (gasstat GC (gas-ionaryionaryMS chromatography-massphase, Duisburg, chromatography (gas phase,phase, 30 chromatography m 30×30Germany) 0.2 mm µ×× m 0.2-0.2mass ID) µµusingm - utilizingmass sspectrometry) ID)pectrometry) a utilizingZB spectrometry)- n1 -columnbutanol n - analyses-butanol analyses (30as analyses them as× 0.25 the were were1internal1 columncolumn runinternal1 column standardon (100%(100% a GCMS standard (100% dimethylpolysiloxanedimethylpolysiloxane for QP2010 dimethylpolysiloxane calibration. for calibration. SE (SHIMADZU The statstat GC Theionaryionary-MS stat GC Duisburg, ionary (gasionary- MSphase,phase, chromatography(gas phase,phase, 3030Germany) chromatography mm ×30×30 0.20.2 mm µµusing××m 0.20.2- massID) µµ am utilizingZB- massID) spectrometry)-1 columnutilizing spectrometry) n--butanol (30 n - butanolm analyses ×as 0.25 the analyses as the wereinternal runwereµminternalinternal on standardid) run a with GCMS standard standardon a the forGCMS QP2010 MS calibration. for for detector QP2010 calibration. calibration.SE (SHIMADZU inSE The electron (SHIMADZU GC The The-MS GC GCimpactDuisburg, (gas--MS Duisburg, (EI) chromatography (gas Germany) mode chromatography Germany) (70 eV).using-mass usinga ZB-- mass s-pectrometry)1 a column ZB spectrometry)-1 column (30 m analyses ×(30× 0.25 m analyses × 0.25µ run3. Experimental on aµminternalwereinternal GCMS id) wereinternalruninternal with standard standard QP2010on Srun theectiona GCMS standard standardon MS aSE for for GCMSdetector QP2010 (SHIMADZU calibration. calibration. for for QP2010 calibration. calibration.in SE electron (SHIMADZU TheSE The ( Duisburg, SHIMADZU GCimpact GC The The--MS GC GC Duisburg, (EI) (gas-MS Germany) mode Duisburg, chromatography (gas Germany) (70 chromatography eV). usingGermany) using a--mass ZB-1 using a -ZB mass scolumnpectrometry)- 1a columnZB spectrometry)-1 (30column (30 m analysesm (30 × 0.25 0.25 analysesm × 0.25m id) µmwere id)µm wererun with id) onA run the witha vacuum GCMS onMS thea detectorGCMS MSQP2010 Fourier detector QP2010 in SE electron transform (SHIMADZU in SE electron ((SHIMADZU impact infrared impact Duisburg,(EI) mode spectrometer Duisburg,(EI) modeGermany)(70 eV). Germany)(70 Bruker eV). using using a Vertex ZB- 1a columnZB 70v--1 column Fourier (30 m (30 ×transform 0.25 m × 0.25 with thewereµm MS id)A runµmwere detector vacuumwith onid) run athewith GCMS on inMS Fourier athe electron GCMS detector QP2010MS transformdetector QP2010 impactin SE electron ((SHIMADZU in SE infraredelectron(EI) (SHIMADZU impact mode impact spectrometerDuisburg, (EI) (70 modeDuisburg, (EI) eV). Germany)mode (70 Bruker eV). Germany) (70 eV). using Vertex using a ZB 70v-- 11a columncolumnZB Fourier-1 column (30(30transform mm (30 × 0.250.25 m × 0.25 µm id)infraredµm with id)id) thewithwith (Buker, MS thethe detector MSMS Milan, detectordetector in Italy). electron inin waselectronelectron impact used impactimpact to(EI) record mode (EI)(EI) modemodethe (70 infraredeV). (70(70 eV).eV). absorption spectra (400–4000 cm−1 infraredµm id)Aid)µm vacuumwithwith (Buker, id)A thethe vacuumwith MSMS Fourier Milan, the detectordetector MS Fourier Italy). transformdetector inin transform waselectronelectron in usedinfrared electron impactimpact to infrared record impact spectrometer (EI)(EI) spectrometermodemodethe (EI) infrared mode (70(70 Bruker eV).eV). (70 absorption Bruker eV). Vertex Vertex 70v spectra Fourier 70v (400 Fourier –transform4000 transform cm−1 3.1. GeneralA vacuumµm id) AInformationµm with vacuum Fourierid)A thewith vacuum MS−1 Fourier the transform detector MS Fourier detector transform in infrared transformelectron in electron infrared impact spectrometer infrared impact spectrometer (EI) spectrometermode (EI) Brukermode (70 Bruker eV). (70 Vertex Bruker eV). Vertex 70v Vertex 70v Fourier Fourier 70v Fourier transform transform transform infrared Ainfraredrange, vacuumA 4 vacuum cm (Buker, Fourier resolution) Milan, Fourier transform Italy).of transform all catalytic was infrared used infrared supports. spectrometer to record spectrometer The the spectra Bruker infrared wereBruker Vertex absorption acquired Vertex 70v utilizing Fourier spectra70v Fourier an (400transform ATR– 4000transformtransform module−1 cm−1 infraredA vacuum (Buker,A−1 vacuum Milan, Fourier Fourier Italy). transform was transform used infrared to infrared record spectrometer the spectrometer infrared Bruker absorption Bruker Vertex Vertex spectra 70v Fourier 70v (400 Fourier– 4000transform cmtransform−1 −1 range,infraredAinfrared 4 vacuumcm (Buker,A resolution)vacuum (Buker, Fourier Milan, Milan,Fourier of Italy). transform all catalytic Italy). transformwas usedwas infrared supports. used to infrared record spectrometer to The record spectrometer thespectra infrared the wereBruker infrared absorptionacquiredBruker Vertex absorption Vertexutilizing 70v spectra Fourier70v spectra an (400 ATR Fourier –− transformtransform4000 (4001 module – transform4000 cm cm −1 −1 (Buker, Milan, infrared Italy). (Buker, was−1 used Milan, to record Italy). was the used infrared to record absorption the infrared spectra absorption (400–4000 spectra cm (400range,–4000−1 4 cm cm−1 infraredrange,infrared (Buker,−1 4 cm (Buker, Milan, resolution) Milan, Italy). of Italy). was all catalytic usedwas used to supports. record to record the The infrared thespectra infrared absorptionwere absorptionacquired spectra utilizing spectra (400 an– (4004000 ATR– 4000 cm module−1 cm−1 range, infrared infrared4 cm (Buker,−1 resolution) (Buker, −1 Milan, Milan, of Italy). all catalytic Italy). was usedwas supports. usedto record toThe record spectrathe infrared the were infrared absorptionacquired absorption utilizing spectra spectra an (400 ATR– 4000 (400 module– cm4000−1−1 cm−1−1 Theinfraredrange,infrared gasrange,infraredinfrared 4chromatography cm (Buker, (Buker, 4 resolution) cm (Buker, (Buker, −1 Milan, Milan, resolution) Milan, Milan, Italy). Italy).of all(G of Italy).catalytic Italy).L wasC) all catalyticanalyses used was supports. usedto supports.recordrecord were to The record the the spectraperformed The infrared infrared the spectra were infrared absorption absorption wereacquiredusing absorption acquired a utilizing GC spectra spectra- utilizing2010 spectra an (400 (400 PlusATR– – an4000 (4004000 moduleATR(SHIMADZU– cm4000 cm module cm range,−1 4 cm−1 resolution) of all catalytic supports. The spectra were acquired utilizing an ATR module resolution) range,of range, 4 allcm catalytic4 resolution) cm−1 resolution) supports. of all ofcatalytic all The catalytic supports. spectra supports. The were spectra The acquired spectra were wereacquired utilizing acquired utilizing an utilizing ATR an ATR module an moduleATR equippedmodule range, range, 4 cm−1−1 4 resolution) cm−1 resolution) of all catalyticof all catalytic supports. supports. The spectra The spectra were acquiredwere acquired utilizing utilizing an ATR an moduleATR module Duisburg range,, Germany range, 4 cm 4 resolution) cm) gas resolution) chromatograph of all catalyticof all catalytic supports. equipped supports. The with spectra The aspectra flame were acquiredwere ionization acquired utilizing detector utilizing an ATR an(FID) moduleATR and module a ZB - with a single reflection diamond ATR crystal (refractive index of 2.4). Each analysis was repeated on 1 column (100% dimethylpolysiloxane stationary phase, 30 m × 0.2 µm ID) utilizing n-butanol as the three samples. internal standard for calibration. The GC-MS (gas chromatography-mass spectrometry) analyses were run on a GCMS QP2010 SE (SHIMADZU Duisburg, Germany) using a ZB-1 column (30 m × 0.25 µm id) with the MS detector in electron impact (EI) mode (70 eV). A vacuum Fourier transform infrared spectrometer Bruker Vertex 70v Fourier transform infrared (Buker, Milan, Italy). was used to record the infrared absorption spectra (400–4000 cm−1 range, 4 cm−1 resolution) of all catalytic supports. The spectra were acquired utilizing an ATR module

Molecules 2017, 22, x 9 of 14

Molecules 2017, 22, 2099 9 of 14 equipped with a single reflection diamond ATR crystal (refractive index of 2.4). Each analysis was repeated on three samples. TheThe surface surface chemical chemical analyse analysess of of the the pristine, pristine, plasma plasma-coated-coated and and used used supports supports were were carried carried out out byby means means of of XPS XPS using using a ausing using a aTheta Theta probe probe spectrometer spectrometer (Thermo (Thermo Electron Electron Corp Corporation,oration, Waltham, Waltham, Massachusetts,MA, USA) spectrometer USA) spectrometer equipped equipped with a monochromatic with a monochromatic Al KaX-ray Al sourceKaX-ray (1486.6 source eV), (1486.6 operating eV), operatingat a spot sizeat a ofspot 300 sizeµm of corresponding 300 µm corresponding to a power to ofa 70power W. Survey of 70 W. (0–1200 Survey eV) (0– and1200 high eV) resolutionand high resolution(C1s, O1s, (C1s, F1s andO1s, Si2p) F1s and spectra Si2p) were spectra recorded were recorded in FAT (fixedin FAT analyzer (fixed analyzer transmission) transmission) mode atmode pass atenergy pass energy of 200 of and 200 100 and eV, 100 respectively. eV, respectively. All spectra All spectra were were acquired acquired at a take-offat a take angle-off angle of 37 of◦ with 37° withrespect respect to the to sample the sample surface. surface. A flood A flood gun wasgun usedwas used to balance to balance the surface the surface charging. charging. The C1s The signal C1s signforal the for hydrocarbon the hydrocarbon component component (285.0 eV) (285.0 was eV) used was as internalused as standard internal standardfor charging for correction. charging correction.For the quantitative For the quantitative analysis, repeated analysis, measurements repeated measurements allowed to register allowed a maximum to register relative a maximum standard relativedeviation standard of about deviation 3%. Atomic of about percentages 3%. Atomic were percentages calculated fromwere thecalculate high resolutiond from the spectra high resolution using the spectraScofield using sensitivity the Scofield factors sensitivity set in the XPS factors data set processing in the XPS software data processing and a non-linear software Shirley and a backgroundnon-linear Shirleysubtraction background algorithm. subtraction algorithm. TheThe morphology morphology of of the the silica silica powders powders was was investigated investigated by by a a SUPRA SUPRA 40 40 field field emission emission scanning scanning electronelectron microscope (FESEM, (FESEM Zeiss, Zeiss ITALIA, ITALIA, Milan, Milan, Italy). Italy SEM). SEM images images were acquired were acquired with a Everhart– with a EverhartThornley–Thornley detector detector at theworking at theworking distance distance in the range in the 7–7.5 range mm, 7–7.5 electron mm, electron acceleration acceleration voltage voltage(extra-high (extra tension,-high tension, EHT) ofEHT) 5 kV, of magnification5 kV, magnification in the in range the range 0.1–50 0.1 KV.–50 The KV. silica The silica powders powders were werefixed fixed on ultrahigh on ultrahigh vacuum vacuum compatible compatible carbon carbon tape tape and sputter-coatedand sputter-coated with with a 20 nma 20 thick nm thick layer layer of Cr ofprior Cr toprior SEM to observation SEM observation utilizing utilizing a turbo-pumped a turbo- sputterpumped coater sputter (model coater Q150T, (model Quorum Q150T, Technologies, Quorum Technologies,Laughton, East Laughton, Sussex, UK) East Sussex, UK) CommercialCommercial solvents andand reagentsreagents of of high high purity purity were were employed. employed. Commercial Commercial (Aldrich, (Aldrich Milan,, Milan, Italy) Italysilica,) gelsilica C8 geland C8 perfluorinated and perfluorinated silica gelsilica 60 gel were 60 200–450were 200 mesh.–450 mesh. Co-solvent Co-solvent water waswater freshly was freshly doubly ® ® doublydistilled distilled before use. before Commercial use. Commercial(Degussa, Essen, (Degussa Germany), Essen Caroat, Germanytriple salt) (2Caroat KHSO 5·tripleKHSO 4 salt·K2 SO (24 , KHSOiodometry5·KHSO 2.864· mmol/g)K2SO4, iodometry was our source 2.86 mmol/g) of potassium was our monoperoxysulfate. source of potassium The soybeanmonoperoxysulfate. oil (iodine number The soybeanof 130 g Ioil2/100 (iodine g) was number provided of 130 by g Valsoia I2/100 (Bologna,g) was provided Italy). by Valsoia (Bologna, Italy ).

3.23.2.. Catalyst Catalyst P Preparationreparation

3.2.13.2.1.. PE PE-CVD-CVD T Treatmentreatment of of C C88-S-Silicailica G Gelel S Supportupport

CommercialCommercial powder powder C C8-8silica-silica gel gel was was treated treated in in the the parallel parallel plate, plate, low low pressure pressure plasma plasma reactor reactor shownshown in in Figure Figure 7.

Figure 7. Schematic of the low pressure plasma reactor utilized for the silica gel treatment. Figure 7. Schematic of the low pressure plasma reactor utilized for the silica gel treatment.

The reactor was composed of a 68 cm long Pyrex glass tube (9 cm internal diameter) closed on one sideThe by reactor a metallic was flange composed sealed of with a 68 cmViton long o-rings Pyrex and glass containing tube (9 cm two internal rectangular diameter) stainless closed steel on electrodesone side by (19 a metallic× 7 cm2). flangeThe lower sealed electrode with Viton was o-rings grounded, and containingwhile the upper two rectangular one was connected stainless steelto a electrodes (19 × 7 cm2). The lower electrode was grounded, while the upper one was connected to

Molecules 2017, 22, 2099 10 of 14 a radio frequency (RF, 13.56 MHz) power supply by means of an L-type matching network. The feed gas flow rate was controlled by electronic mass flow controllers (MFC), and the pressure was measured by a capacitance vacuum gauge. A rotative pump and a manual throttle valve allowed for a constant working pressure during the surface modification processes. During the PE-CVD experiments, the plasma was fed with hexafluoropropene (C3F6) at a flow rate of 30 sccm, and the input power was fixed at 50 W. The grafting processes was performed with CF4 at a flow rate of 20 sccm and 20 W input power. Plasma treatments were carried out on 1.0 g of catalytic support distributed over a flat stainless steel container positioned on the grounded electrode.

3.2.2. Preparation of Supported Catalysts by Impregnation

The composite materials were prepared by impregnation of C8 silica gel supports A-C with an appropriate quantity of methyl perfluoroheptyl ketone (2). Typically, ketone (2) was dissolved in 2.5 mL of perfluorohexane and 1 g of finely ground silica support was added to this solution. The resulting dispersion was stirred overnight at room temperature and afterwards the solid was separated by centrifugation and dried. The GC/MS analysis of the supernatant solution shows that ketone (2) is completely absorbed. Silica gel composites thus obtained, labelled as 2-(A), 2-(B) and 2-(C), were typically loaded with of ketone (2) in the range of 5 × 10−2–15 × 10−2 mmol per 100 mg of support.

3.3. Epoxidation Procedure The oxidation of substrates was carried out in a batch reactor operating at a fixed temperature (25 ◦C). In a typical experiment, the substrate (3–9, 1 mmol) was added to a suspension of catalyst 2-(C) (100 mg) in CH3CN (5 mL) and aqueous phosphate buffer (5 mL, 0.8 M, pH 7.5), also containing Na2EDTA (4.0 g/L ca.) to sequester any traces of heavy metals. Under vigorous stirring, solid caroate (1.32 g, 4 mmol) was added portionwise over 10–30 min, and the resulting mixture was allowed to react at 20–25 ◦C for an appropriate amount of time (1–12 h). The reaction progress was monitored as follows: aliquots (0.2 mL), periodically withdrawn from the reaction flask, were quickly filtered through a cotton plug to remove the solid catalyst, diluted to 2 mL with acetonitrile containing the internal standard, and analysed by GC. In each experiment, the values were measured at various reaction times; data from two or more independent runs were averaged (estimated error is ±3%). Upon reaction completion, the reaction mixture was partitioned between water (10 mL) and CH2Cl2 (10 mL) then filtered through a Buchner funnel to remove the suspended catalyst. The organic layer was separated and dried over MgSO4, and the solvent was removed under reduced pressure to afford the crude products identified by GC-MS analysis. In most of the cases, pure epoxide products (95–97%, GC) were obtained by flash chromatography (silica gel, n-hexane/Et2O). The isolated yields were determined by considering the total amount of substrate, usually 1 mmol. The products were characterized by 1H-NMR, GC-MS, and identified by comparison with this data with the spectral data of authentic samples. (E)-β-Methylstyrene oxide (3a)[34,35]. (107 mg (0.80 mmol), 80% isolated yield, GC purity 95%): colorless oil, bp 93–96 ◦C/16 mmHg. Cyclohexene oxide (4a)[35,36]. (92 mg (0.94 mmol, 94% isolated yield GC purity 95%) colorless oil, bp 129–130 ◦C. 3-Hydroxycyclohex-1-ene oxide (5a)[28]. (87 mg (0.76 mmol), 76% isolated yield, GC purity 95%): colorless oil, syn/anti ratio 60/40 (GC). Cyclohex-2-enone (5b)[28]. (12 mg (0.13 mmol), 13% isolated yield, GC purity 95%). 1-Octene oxide (6a)[35]. (16 mg (0.12 mmol), 13% isolated yield, GC purity 97%; recovered 1-octene (6) substrate: 94 mg (0.84 mmol)): colorless oil, bp 62–65 ◦C/17 mmHg. Molecules 2017, 22, 2099 11 of 14

Hexane-3,4-dione (8a)[29]. (108 mg (0.95 mmol), 95% isolated yield, GC purity 95%): colorless oil, 131–133 ◦C. Methyl oleate oxide (10a)[32,37]. (260 mg, 83% isolated yield, GC purity 90%): colorles oil. Epoxidized soybean oil (ESO) (11a)[32,38]. see 1H-NMR in Supplementary Materials.

3.4. Recycling Experiments The stability of catalyst 2-(C) was studied in a series of recycling experiments. In the first run, substrate 3 (1 mmol) was added to a suspension of 100 mg of catalyst 2-(C) in CH3CN (5 mL) and aqueous phosphate buffer (5 mL, 0.8 M, pH 7.5), also containing Na2EDTA (4.0 g/L ca.) to sequester any trace of heavy metals. Under vigorous stirring, solid caroate (1.32 g, 4 mmol) was added portionwise over 10–30 min, and the resulting mixture was allowed to react at 20–25 ◦C for 4 h. After completion of reaction, the solid residue was decanted and supernatant aqueous buffer was carefully removed washing with little aliquots of fresh solvent. Then, fresh buffer and reagents were added to the solid for a new run. After each cycle, aqueous supernatant was treated for evaluating conversions by partition between water (10 mL) and CH2Cl2 (10 mL). The organic layer was separated and dried over MgSO4, and the solvent removed under reduced pressure to afford the crude products identified by GC-MS analysis.

3.5. TON Measurement To evaluate the whole catalyst robustness, epoxidation of trans-β-methyl styrene (3) was carried out with excess amounts of alkene and monitored until deactivation. In a typical experiment, to 2.50 g (21.186 mmol) of substrate (3) and 0.100 g (bearing adsorbed 9.7 × 10−2 mmol of ketone (2) of catalyst 2-(C) in acetonitrile (5 mL) and aqueous buffer (10 mL, pH 8), caroate (1.40 g, 4 mmol) was added portion wise together with further buffer aliquots (5 mL) during 48 h. After addition of a total amount of 33.0 g (95 mmol) of caroate, no further progress in alkene conversion (95%, ca. 48 h) was detected (GC monitoring). Based on the substrate amount converted (20 mmol) and the catalyst loading (0.097 mmol) the calculated turnover number (TON) value was 206.

4. Conclusions In summary, a novel heterogeneous catalyst for epoxidation reactions was prepared by the modification of the surface of commercial C8-silica gel with non-equilibrium low pressure plasma fed with hexafluoropropene (C3F6). The successive impregnation with ketone (2) of the C8-silica gel coated with a thin perfluorinated film, afforded the organocatalyst 2-(C) suitable for dioxirane-mediated epoxidations with caroate (KHSO5) as a green oxidant. This method extends our previous approach on the Merrifield’s resin [17] simplifying the catalyst preparation and widening the substrate scope. Results, coupled with data from chemical characterization of the plasma-treated silica gel, suggest a number of strong points of this protocol: (i) an increased reactivity of ketone (2) adsorbed onto the perfluorinated matrix, (ii) an activating effect of the support towards caroate by means of a multiple weak hydrogen-bond network, with generation of singlet oxygen; (iii) the possibility of performing two different kinds of olefin transformations (either epoxidation or double bond cleavage) by means of the simple choice of starting reagents; (iv) the opportunity of extending epoxidation to unsaturated fatty esters like methyl oleate and the triglyceride soybean oil, thus producing bulk chemicals of great practical importance such as ESO; (v) the faculty of recycling the catalyst up to five times and (vi) the high catalyst activity with a total TON value of 206 ca. Finally, the generation of singlet oxygen also suggests the possibility of using these new perfluorinated silica gel matrixes for applications in photodynamic therapy [39]. All these significant benefits are in accordance with the rules of green chemistry and can provide to the method flexibility and a good potential for industrial applications [40]. Molecules 2017, 22, 2099 12 of 14

Supplementary Materials: The following are available online, Table S1: XPS characterization of Catalyst 2-(C); Figure S1: XPS C1s region, curve fitting for 2-(C); Figure S2: XPS O1s region, curve fitting for catalyst 2-(C); Figure S3: 1H-NMR spectrum of ESO. Acknowledgments: The research was supported by CNR ICCOM and Regione Puglia MIUR PON Ricerca e Competitività 2007–2013 Avviso 254/Ric. del 18/05/2011, Project PONa3 00369 “Laboratorio SISTEMA”. Commercial ® Caroat triple salt (2 KHSO5·KHSO4·K2SO4) was kindly supplied by Peroxide-Chemie (Degussa, Germany). Author Contributions: L.D. and F.Fracassi conceived and designed the experiments; N.D.V. performed the experiments; C.F. and A.N. analyzed the data; F.Fanelli performed SEM analysis; L.D. and F.Fracassi wrote the paper. Conflicts of Interest: The authors declare no conflict of interest.

References

1. Berger, R.; Resnati, G.; Metrangolo, P.; Weberd, E.; Hulliger, J. Organic fluorine compounds: A great opportunity for enhanced materials properties. J. Chem. Soc. Rev. 2011, 40, 3496–3508. [CrossRef][PubMed] 2. Berkessel, A.; Adrio, J.A.; Hüttenhain, D.; Neudörfl, J.-M. Unveiling the “Booster Effect” of Fluorinated Alcohol Solvents: Aggregation-Induced Conformational Changes and Cooperatively Enhanced H-Bonding. J. Am. Chem. Soc. 2006, 128, 8421–8426. [CrossRef][PubMed] 3. Li, W.; Fuchs, P.L. A New Protocol for in Situ Dioxirane Reactions: Stoichiometric in Oxone and Catalytic in Fluorinated Acetophenones. Org. Lett. 2003, 5, 2853–2856. [CrossRef][PubMed] 4. Ghaffarzadeh, M.; Ahmadi, M. Catalytic application of fluorous silica gel in Fries rearrangement. J. Fluor. Chem. 2014, 160, 77. [CrossRef] 5. Bernini, R.; Cacchi, S.; Fabrizi, G.; Forte, G.; Niembro, S.; Petrucci, F.; Pleixats, R.; Prastaro, A.; Sebastian, R.M.; Soler, R.; et al. Phosphine-Free Perfluoro-Tagged Palladium Nanoparticles Supported on Fluorous Silica Gel: Application to the Heck Reaction. Org. Lett. 2008, 10, 561–564. [CrossRef][PubMed] 6. Hong, M.; Xiao, G. Bis(perfluorooctanesulfonyl)imide supported on fluorous silica gel: Application to protection of carbonyls. J. Fluor. Chem. 2012, 140, 121–126. [CrossRef] 7. Bernini, R.; Cacchi, S.; Fabrizi, G.; Forte, G.; Niembro, S.; Petrucci, F.; Pleixats, R.; Prastaro, A.; Shafir, A.; Vallribera, A. Perfluoro-Tagged Gold Nanoparticles Immobilized on Fluorous Silica Gel: A Reusable Catalyst for the Benign Oxidation and Oxidative Esterification of Alcohols. ChemSusChem 2009, 2, 1036–1040. [CrossRef][PubMed] 8. Harald, S.; Petra, H.; Lechner, R.; Burkhard, K. Photooxidation of Benzyl Alcohols with Immobilized Flavins. Adv. Synth. Catal. 2009, 351, 163–174. [CrossRef] 9. Annese, C.; D’Accolti, L.; Filardi, R.; Tommasi, I.; Fusco, C. Oxidative cleavage of lactams in water using dioxiranes: An expedient and environmentally-safe route to ω-nitro acids. Tetrahedron Lett. 2013, 54, 515–517. [CrossRef] 10. Annese, C.; Abbrescia, D.I.; Catucci, L.; D’Accolti, L.; Denora, N.; Fanizza, I.; Fusco, C.; La Piana, G. Site-dependent biological activity of valinomycin analogs bearing derivatizable hydroxyl sites. J. Pept. Sci. 2013, 19, 751–757. [CrossRef][PubMed] 11. Calò, V.; Nacci, A.; Fiandanese, V. Copper-mediated regio- and enantio-selective cross-coupling of heterocyclic allyl sulphides with organomagnesium compounds: A case of 1,7-relative stereogenesis. Tetrahedron 1996, 52, 10799–10810. [CrossRef] 12. Cosma, P.; Catucci, L.; Fini, P.; Dentuto, P.L.; Agostiano, A.; Angelini, M.; Scolaro, L.M. Tetrakis(4-pyridyl) porphyrin supramolecular complexes with cyclodextrins in aqueous solution. Photochem. Photobiol. 2006, 2, 563–569. [CrossRef][PubMed] 13. Annese, C.; D’Accolti, L.; Dinoi, A.; Fusco, C.; Gandolfi, R.; Curci, R. Concerning the Reactivity of Dioxiranes. Observations from Experiments and Theory. J. Am. Chem. Soc. 2008, 130, 1197–1204. [CrossRef][PubMed] 14. Denmark, S.E.; Forbes, D.C.; Hay, D.S.; De Pue, J.S.; Wilde, R.G. Catalytic Epoxidation of Alkenes with Oxone. J. Org. Chem. 1995, 60, 1391–1407. [CrossRef] 15. Cozzi, F. Immobilization of Organic Catalysts: When, Why, and How. Adv. Synth. Catal. 2006, 348, 1367–1390. [CrossRef] Molecules 2017, 22, 2099 13 of 14

16. D’Accolti, L.; Annese, C.; De Riccardis, A.; De Giglio, E.; Cafagna, D.; Fanelli, F.; Fusco, C. Dioxirane-Mediated Heterogeneous Epoxidations with Potassium Caroate: A Solid Catalyst Bearing Anchored Ketone Moieties. Eur. J. Org. Chem. 2012, 24, 4616–4621. [CrossRef] 17. De Vietro, N.; Annese, C.; D’Accolti, L.; Fanelli, F.; Fusco, C.; Fracassi, F. A new Synthetic Approach to Oxidation Organocatalysts Supported on Merrifield Resin using Plasma-Enhanced Chemical Vapor Deposition. Appl. Catal. A Gen. 2014, 470, 132–139. [CrossRef] 18. Fanelli, F.; Fracassi, F.; d’Agostino, R. Fluorination of polymers by means of He/CF4-Fed atmospheric pressure glow dielectric barrier discharges. Plasma Process. Polym. 2008, 5, 424–432. [CrossRef] 19. Fanelli, F.; Fracassi, F.; d’Agostino, R. Deposition and etching of fluorocarbon thin films in atmospheric

pressure DBDs fed with Ar-CF4-H2 and Ar-CF4-O2 mixtures. Surf. Coat. Technol. 2010, 204, 1779–1784. [CrossRef] 20. Legros, J.; Crousse, B.; Bonnet-Delpon, D.; Bégué, J.-P. Design of fluoroketones as efficient reagents for epoxidation reactions in hexafluoropropan-2-ol. Tetrahedron 2002, 58, 3993–4324. [CrossRef] 21. Fanelli, F.; Fracassi, F. Atmospheric pressure non-equilibrium plasma jet technology: general features, specificities and applications in surface processing of materials. Surf. Coat. Technol. 2017, 322, 174–201. [CrossRef] 22. Adam, W.; Kazakov, D.V.; Kazakov, V.P. Singlet-Oxygen Chemiluminescence in Peroxide Reactions. Chem. Rev. 2005, 105, 3371–3387. [CrossRef][PubMed] 23. Pace, A.; Pierro, P.; Buscemi, S.; Vivona, N.; Clennan, E.L. Photooxidations of Alkenes in Fluorinated Constrained Media: Fluoro-organically Modified NaY as Improved Reactors for Singlet Oxygen “Ene” Reactions. J. Org. Chem. 2007, 72, 2644–2646. [CrossRef][PubMed] 24. Harbour, J.R.; Issler, S.L. Involvement of the Azide Radical in the Quenching of Singlet Oxygen by Azide Anion in Water. J. Am. Chem. Soc. 1982, 104, 903–905. [CrossRef] 25. Hajimohammadi, M.; Safari, N. Photooxygenation of alkenes by molecular oxygen in the presence of porphyrins and chlorin sensitizers under visible light irradiation. J. Porphyr. Phthalocyanines 2010, 14, 639–645. [CrossRef] 26. Croxtall, B.; Fawcett, J.; Hope, E.G.; Stuart, A.M. Synthesis and coordination chemistry of perfluoroalkyl-derivatised β-diketonates. J. Fluor. Chem. 2003, 119, 65–73. [CrossRef] 27. Chen, R.; Silverstein, M.S. Plasma copolymerization: Hexafluoropropylene and a nonpolymerizable gas. J. Polym. Sci. Part A Polym. Chem. 1996, 34, 207–216. [CrossRef] 28. Murray, R.W.; Singh, M.; Williams, B.L.; Mocrieff, H.M. Diastereoselectivity in the Epoxidation of Substituted Cyclohexenes by Dimethyldioxirane. J. Org. Chem. 1996, 61, 1830–1841. [CrossRef][PubMed] 29. Antoniotti, S.; Alezra, N.; Fernandez, X.; Duñach, E. Catalytic epoxide oxidation: A novel access to flavouring and odorant α-diketones. J. Flavour Fragr. 2004, 19, 373–378. [CrossRef] 30. D’Accolti, L.; Fiorentino, M.; Fusco, C.; Crupi, P.; Curci, R. Selective oxidation of acetylenic 1,4-diols with Dioxiranes in comparison with the Methyltrioxorhenium-Hydrogen Peroxide oxidant. Tetrahedron Lett. 2004, 45, 8575–8578. [CrossRef] 31. Akiyama, T.; Enomoto, Y.; Shibamoto, T. A new method of pyrazine synthesis for flavor use. J. Agric. Food Chem. 1978, 26, 1176–1179. [CrossRef] 32. Pantone, V.; Annese, C.; Fusco, C.; Fini, P.; Nacci, A.; Russo, A.; D’Accolti, L. One-Pot Conversion of

Epoxidized Soybean Oil (ESO) into Soy-Based Polyurethanes by MoCl2O2 Catalysis. Molecules 2017, 22, 333. [CrossRef][PubMed] 33. Fenollar, O.; García, D.; Sánchez, L.; López, J.; Balart, R. Optimization of the curing conditions of PVC plastisols based on the use of an epoxidized fatty acid ester plasticizer. Eur. Polym. J. 2009, 45, 2674–2684. [CrossRef] 34. Curci, R.; D’Accolti, L.; Fiorentino, M.; Rosa, A.M. Enantioselective Epoxidation of Unfunctionalized Alkenes using Dioxiranes Generated in Situ. Tetrahedron Lett. 1995, 32, 5831–5834. [CrossRef] 35. Ankudey, E.G.; Olivo, H.F.; Peeples, T.L. Lipase-mediated epoxidation utilizing urea–hydrogen peroxide in ethylacetate. Green Chem. 2006, 8, 923–926. [CrossRef] 36. Wiles, C.; Hammond, M.J.; Watts, P. The development and evaluation of a continuous flow process for the lipase-mediated oxidation of alkenes. Beilstein J. Org. Chem. 2009, 5, 27. [CrossRef][PubMed] 37. Bunker, S.P.; Wool, R.P. Synthesis and characterization of monomers and polymers for adhesives from methyl oleate. J. Polym. Sci. Part A Polym. Chem. 2002, 40, 451–458. [CrossRef] Molecules 2017, 22, 2099 14 of 14

38. Dai, H.; Yang, L.; Lin, B.; Wang, C.; Shi, G. Synthesis and characterization of the different soy-based polyols by ring opening of epoxidized soybean oil with methanol, 1,2-ethanediol and 1,2-propanediol. J. Am. Oil Chem. Soc. 2009, 86, 261–267. [CrossRef] 39. Ghogare, A.A.; Greer, A. Using Singlet Oxygen to Synthesize Natural Products and Drugs. J. Org. Chem. 2012, 77, 4557–4565. [CrossRef][PubMed] 40. Sheldon, R.A. Green and sustainable manufacture of chemicals from biomass: State of the art. Green Chem. 2014, 16, 950–963. [CrossRef]

Sample Availability: Samples of the compounds are not available from the authors.

© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).