Social Media

Alkene-Zipper Catalyzed Selective and Remote Retro-ene Reaction of Alkenyl Cyclopropylcarbinol

Alkene-Zipper Catalyzed Selective and Remote Retro-ene Reaction of Alkenyl Cyclopropylcarbinol
of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657   Very Important Publication Alkene-Zipper Catalyzed Selective and Remote Retro-eneReaction of Alkenyl Cyclopropylcarbinol Jeffrey Bruffaerts, +a Alexandre Vasseur, +a and Ilan Marek a, * a The Mallat Family Laboratory of Organic Chemistry, Schulich Faculty of Chemistry, Technion-Israel Institute of Technology, Haifa 32000, IsraelE-mail: + Equal contribution Received: November 20, 2017; Revised: January 16, 2018; Published online: February 13, 2018Supporting information for this article is available on the WWW under Abstract:  Reminiscent of biological systems, theincreasingly popular concept of remote activationallows a greater strategic synthetic flexibility for thedevelopment of novel synthetic organic method-ologies. In this Communication, we report thatcommercially available ruthenium (II)-based “al-kene zipper catalyst” enables the selective trans-formation of a large variety of   w -alkenyl cyclo-propylcarbinols into stereodefined unconjugated( E  )-acyclic aldehydes bearing a quaternary stereo-center through an isomerization followed by aretro-ene reaction. To unravel this peculiar catalyticproperty of the “alkene zipper” and shed some lighton the mechanism, a series of control experimentswere performed. Keywords:  Remote functionalization; retro-ene;Ruthenium; Mechanistic investigation; Catalyticalkene isomerization.The strategy of remote functionalization, consisting inan initial interaction of a functional group promotinga selective reaction at a distal position, has becomeincreasingly important in organic synthesis. Firstmentioned by Breslow in the early 70s for biomimetictransformations, [1] and later used by Schwarz in gas-phase chemistry, [2] this concept has only latelyemerged as an srcinal and powerful methodology forsynthetic transformations, thanks to the developmentof novel transition-metal based systems. [3] This para-digm shift has allowed the development of newmethodologies through srcinal approaches. Amongthem, the distant activation of an “unactivated posi-tion” strongly modifies our perception of organictransformations which was mostly concentrated ondirect transformation of the most chemically reactiveposition. For instance, the transition metal-mediatedisomerization of a double-bond potentially representsa unique opportunity to transfer chemical informationover an aliphatic chain. [4] Coupling the transpositionof the unsaturation with a subsequent remote site-selective functionalization would clearly illustrate thepower of using a mixture of positional and geometricalolefin isomers into a single well-defined product.Furthermore, the development of such tandem reac-tions, where all events are taking place in a single-potoperation [5] would answer all the requirements of modern organic synthesis. [6–8] In the last few years, ourresearch group has been interested in the selectivering-opening of strained carbocycles to reveal quater-nary carbon centers in acyclic systems [9] and in thecontext of our studies on remote functionalization, wewere interested to initiate a reaction at a distantposition to subsequently trigger a selective carbon-carbon cleavage at a different position of the molec-ular backbone. [10] For instance, we previously showedthat stoichiometric amount of the Negishi reagent(Cp 2 ZrC 4 H 8 ) [11] could efficiently promote a Zr-migra-tion along the hydrocarbon chain [12] of an  w -enecyclopropane to induce the ring-strain release of thethree-membered ring and yield the  s  - alkyl- p  -allylzir-conocene complex. [13] This stereodefined bismetallicintermediate can then be modularly and stereoselec-tively functionalized with two different electrophilesto afford highly valuable acyclic molecular frame-works bearing two stereogenic centers with excellentlevels of diastereocontrol (Scheme 1a). [13] Alterna-tively, if a leaving group is judiciously introduced inthe carbon skeleton, the zirconocene complex inducesa migration of the double bond followed by a selectivefragmentation of the starting  w -alkenyl cyclopropylcarbinol methyl ether to afford, after acetolysis, the COMMUNICATIONS DOI: 10.1002/adsc.201701481  Adv. Synth. Catal.  2018 ,  360 , 1389–1396 1389  2018 Wiley-VCH Verlag GmbH &  Co. KGaA, Weinheim  123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657   corresponding acyclic skipped dienes in excellent yield(Scheme 1b). [13] Noteworthy, Micalizio also earlierreported similar approach to cleave vinyl cyclopropylcarbinol derivatives employing a titanium mediatedreagent to access various unconjugated dienes bearingsterocenters (Scheme 1c). [14] These aforementionedreactions have the disadvantage of employing stoi-chiometric amounts of transition metal, and in the lasttwo cases, eliminating a chemical handle. Instead of displacing this functionality, and based on the Pd-catalyzed enantioselective Heck arylation using aredox-relay strategy (Scheme 1d), [15] we could developa catalytic process triggering the remote cyclopropaneunfolding to produce useful linear carbonyl groupswith the formation of two contiguous stereocentersincluding a quaternary carbon center (Scheme 1e). [16,17] An alternative mode of concomitant unfolding of athree-membered ring could also be produced by theuncatalyzed retro-ene reaction of   cis -(2-vinylcyclopro-py1)carbinol (1,5-hydrogen shift, Scheme 1f). How-ever, this reaction was shown to proceed only at ratherhigh temperatures (300 8 C at 105 torr) without anyselectivity. [18] To the best of our knowledge, such retro-ene fragmentation has never been investigated undermilder conditions through the use of a catalyst(Scheme 1g). In this communication, we wish to reporta synthetic transformation fulfilling this criterionemploying the commercially available “alkene zip-per” [19c,20] catalyst ( 4 ), that selectively converts alkenylcyclopropylcarbinols  1  and  3  into linear and uncon- jugated ( E  )-enaldehydes  2 . In the latter case, meetingthis challenge results in a four-events cascade, namely( i ) isomerization of a double bond across long-rangehydrocarbon distances, ( ii ) selective C  C bond cleav-age, ( iii ) selective formation of a stereodefined olefinand ( iv ) preservation of the stereointegrity of thequaternary stereocenter. One clear advantage of thisapproach is that diastereo- and enantiomerically purestereodefined polysubstituted cyclopropylcarbinol de-rivatives  1  and  3  are easily accessible in only 2chemical steps from commercially available al-kynes, [16,21] suggesting that the proposed combinedisomerization – retro-ene reaction would represent apowerful strategy to prepare potentially valuableacyclic molecular fragments featuring the challengingquaternary carbon stereocenter, an aldehyde moietyand a stereodefined double bond ( 2 ) in only threecatalytic steps.For instance, when  1a  (R 1 = R 3 = Me, R 2 = Bu) wastreated with 5 mol% of   4  at 75 8 C in a polar solventsuch as 3-pentanone for 30 min, the aldehyde  2a  wasobtained in 92% yield as a unique isomer ( E  : Z   99:1)as described in Scheme 2. When the isotopicallylabeled starting material  1b d2  was treated under thesame condition,  2b d2  was similarly produced with onedeuterium atom present at the allylic position, sustain-ing a metal-catalyzed 1,5-hydride transfer (see pro-posed mechanism on Scheme 3).Before further proceeding and eventually extend-ing the scope of this retro-ene reaction, we wanted to Scheme 1.  Previous studies unfolding cyclopropyl rings ( a – c ),palladium-catalyzed remote functionalization ( d ,  e ) thermalretro-ene reaction ( 1f  ) and proposed metal-catalyzed retro-ene ( 1g ) combined with an isomerization reaction ( 1h ). Scheme 2.  Ru-catalyzed retro-ene reaction. COMMUNICATIONS  Adv. Synth. Catal.  2018 ,  360 , 1389–1396 1390  2018 Wiley-VCH Verlag GmbH &  Co. KGaA, Weinheim  123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657   confirm that the same catalyst  4  would also be thecatalyst of choice to promote the isomerization of thedouble bond of allylcyclopropyl carbinol  3a  in combi-nation with the retro-ene reaction (as described inScheme 1f). We started our study by screening variouscatalytic systems known to promote the conversion of alkenyl alcohols into saturated carbonyl compoundsthrough double-bond isomerization across long-rangemolecular distances [19] such as Ru(II), Rh(I) andPd(II)-based catalysts (Table 1). For instance, RuHCl(CO)(PPh 3 ) 3 , which was shown by Ryu to promoteboth migration of the double bond and subsequentreaction upon the resulting ruthenium enol allylspecies, [22] catalyzed the desired transformation toafford  2a  in moderate yield and selectivity (Table 1,entry 1).Other metal hydride complexes showed almostnone or relatively poor activity (Table 1, entries 2–8);HRh(PPh 3 ) 3  afforded a significant amount of   2a  butwith a disappointing 1:1  E/Z   ratio. Regarding palla-dium hydride catalysts, which were  in situ  generatedfrom Pd(II) precatalysts Pd(phen)MeCl or Pd(dcpe)MeCl, NaBAr f  and cyclohexene, respectively reportedby Kochi [23] and Mazet [17e] , they also failed to affordthe desired product ( 2a ), leading instead to oligomeri-zation side-reactions (Table 1, entries 7–8). To ourdelight, the same “alkene zipper” catalyst ( 4 ) pre-viously discussed, [20] for the retro-ene reaction(Scheme 2) was found to be the only catalyst promot-ing a fast, selective and efficient isomerization coupledwith the subsequent retro-ene reaction in our exper-imental conditions. Known to rapidly isomerize ole-fins, via an  allyl   mechanism (or 1,3-H shift), [20] thecatalyst ( 4 ) smoothly enabled the conversion of   3a into  2a  as single  E  -isomer under relatively mildconditions, namely 5 mol% catalyst at 75 8 C using 3-pentanone as solvent (Table 1, entry 9). Noteworthy,performing the reaction in acetone at reflux (57 8 C)led to a sluggish reaction, whereas increasing thetemperature (90 8 C) slightly decreased the  E/Z   ratio of  2a  but without any subsequent re-isomerization of thedouble bond (Table 1, entries 10, 11, respectively).Using polar solvents such as DMF resulted in a poorconversion and selectivity (Table 1, entry 12), while noreactivity was observed in DMSO (Table 1, entry 13).Only chlorinated solvents such as 1,2-dichloroethane(DCE) or 1,1,2,2-tetrachloroethane (TCE) appearedto be suitable solvents although the reaction exhibitedslower kinetics in comparison with the reactionsperformed in 3-pentanone and therefore requiredsignificantly longer reaction times (Table 1, entries 14and 15). Conducting the reactions in less polar solvent Scheme 3.  Proposed mechanism. COMMUNICATIONS  Adv. Synth. Catal.  2018 ,  360 , 1389–1396 1391  2018 Wiley-VCH Verlag GmbH &  Co. KGaA, Weinheim  123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657 such as toluene and 2-MeTHF only afforded traces of  2a  (Table 1, entries 16 and 17). These results some-what suggested an indirect and still unclear role of thesolvent 3-pentanone on the enhancement of thecatalytic property of the ruthenium metal complex insolution. To confirm that the transformation of   3a  into 2a  proceeded first by the formation of the vinyl-cyclopropane intermediate  1  followed by the retro-enerearrangement, the reaction was monitored by  1 HNMR (Figure 1a). At room temperature, neither traceof substrate  3a  nor ring-opening product  2a  wasdetected indicating that the isomerization step fromthe allyl cyclopropylcarbinol ( 3a ) into the ( E  )-vinylcyclopropyl carbinol ( 1a ) proceeded in few minutes atroom temperature. Furthermore, the conversion of   1a into  2a  could only be observed upon heating thereaction mixture around 75 8 C, strongly suggestingthat the ring-opening event might be the ratedetermining step of the reaction. We evaluated thebarrier of activation for this key step to be of 13 kcalmol  1 at 298 K in deuterated 1,1,2,2-tetra-chloroethane (Figure 1b). This result sharply contrastswith the high barrier activation that is required for thethermal-induced retro-ene reaction (300 8 C, see Sche-me 1f). [18] An isotopic labelling experiment (Figure 1c) en-abled to crucially understand the ring-opening mode.Indeed,  3a d2  was smoothly converted into  2a d2 , inwhich one deuterium atom was incorporated in theallylic position, similarly to what was observed for theconversion of   1b d2  into  2b d2  (Scheme 2). The relativelysmall kinetic isotopic effect of the hydrogens in  a -position of the alcohol (KIE  1.1) does not allow usto propose with certainty a mechanistic pathwayalthough the Ru-promoted H-abstraction has beendescribed to proceed through an inner-sphere protonabstraction followed by a proton transfer to theligand. [24] Furthermore, to assess the role of the alcohol inthe mechanism, we studied the effect of its relativeconfiguration towards the allylic chain. To this end, wesynthesized the  anti -derivative of   3a  ( 3a anti  dr  97:3:0:0). [21a] Our experiment highlighted that theconversion of   3a anti  into  2a  required higher temper-atures (95 8 C) and prolonged reaction times (13 h) toonly afford 63% of aldehyde  2a  accompanied by un-fragmented alcohols  1 syn  and  1 anti .Lastly, Figure 1d illustrates that the reaction ratedramatically drops when  3a anti  is used instead of   3a syn as the conversion to  2a  from the former was evaluatedto be 13 times slower than the latter (Figure 1d).Despite the fact that the overall yield was lower forthe transformation of   3a anti , into  2a , both substrates( 3a anti  and  3a syn ) converge to the formation of thesame ( E  )-isomer  2a . This last result might indicatethat the fragmentation step proceeded through acommon intermediate.   Table 1.  Optimization conditions for the catalytic conversionof model substrate  3a  into  2a .Entry Catalytic system Conditions [a] Yield [b] E  : Z  [c] 1 HRuCl(CO)(PPh 3 ) 3 (10 mol%)Toluene,90 8 C, 13 h63 75:252 HRuCl(PPh 3 ) 3 (10 mol%)Toluene,90 8 C, 13 h < 5 –3 RuH 2 (CO) 2 (PPh 3 ) 2 (10 mol%)Toluene,90 8 C, 13 h8 60:404 Ru-MACHO(10 mol%)Toluene,90 8 C, 13 h0 –5 HRh(CO)(PPh 3 ) 3 (10 mol%)Toluene,90 8 C, 13 h [e] < 5 –6 HRh(PPh 3 ) 3  (10 mol%) Toluene,90 8 C, 13 h [e] 26 50:507 Pd(dcpe)MeCl(5 mol%),NaBAr f  (5.5 mol%), cy-clohexene (50 mol%)DCE, 85 8 C,20 h [f] 0 [f] –8 Pd(phen)MeCl(2 mol%),NaBAr f  (2.5 mol%), cy-clohexene (50 mol%)DCE, 85 8 C,20 h [f] < 5 [f] –9  4  (5 mol%) Et 2 CO,75 8 C, 30 min89 98:210  4  (5 mol%) Acetone,57 8 C, 13 h9 97:311  4  (5 mol%) Et 2 CO, 90 8 C 71 95:512  4  (5 mol%) DMF, 75 8 C,20 h25 [d] 75:2513  4  (5 mol%) DMSO,75 8 C, 20 h014  4  (5 mol%) 1,1,2,2-TCE,75 8 C, 20 h95 [d] 98:215  4  (5 mol%) 1,2-DCE,75 8 C, 20 h95 [d] 98:216  4  (5 mol%) 2-MeTHF  < 5 [d] 17  4  (5 mol%) Toluene  < 5 [d][a] Performed on 0.8 mmol scale unless otherwise noted. [b] Isolated yield after column chromatography. [c] Determined by  1 H NMR analysis. [d] Conversion determined by GC and NMR analyses. [e] Performed on 0.4 mmol scale. [f] Complex reaction mixture Ru-MACHO = {bis[2-(diphe-nylphosphino)ethyl]amine}carbonylchlorohydridoruthe-nium(II). dcpe = bis(dicyclohexylphosphino)ethane.phen = 1,10-phenanthroline. NaBAr f  =  tetrakis(3,5-bis(tri-fluoromethyl)phenyl)borate. DCE = 1,2-dichloroethane;Et 2 CO = diethyl ketone; 2-MeTHF = 2-methyltetrahydro-furan; 1,1,2,2-TCE = 1,1,2,2-tetrachloroethane; DMF = N,N  -dimethylformamide, DMSO = dimethylsulfoxide. COMMUNICATIONS  Adv. Synth. Catal.  2018 ,  360 , 1389–1396 1392  2018 Wiley-VCH Verlag GmbH &  Co. KGaA, Weinheim  123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657 Based on these mechanistic investigations (Fig-   ure 1), we suggest a stepwise sequence involvingalkene isomerization followed by a retro-ene pro-moted ring-fragmentation (Scheme 3). Coordinationof the alkene  3a  to  4  would first afford the complex  I .The terminal olefin complex ( I ) would then beconverted into the more thermodynamically ( E  )-1,2-alkene isomer ( II ). This single step, studied bothexperimentally [20] and theoretically, [24b,c] was shown toformally proceed through a 1,3-hydrogen shift, thanksto a hydrogen shuttle with the ligand, enabling a fastalkene isomerization to stereoselectively yield the( E  )-isomer. Intermediate  II  would then undergo aruthenium-promoted concerted C 5 -H hydrogen ab-straction with cleavage of the proximal C  C bond of the three-membered ring to afford the allyl rutheniumhydride complex  III . Eventually,  2a  and  4  would bereleased following the insertion of the double bondinto the Ru  H bond and tautomerization. If   1  isemployed, we propose that an interaction between thealcohol and the cationic metal center might favor theapproach of the complex towards the stericallyencumbered polysubstituted cyclopropane to promotethe retro-ene reaction (formation of   III  from  II  andthen release of   2a ). In the case of the conversion of  3a anti , we suggest that the metal complex  IV  would Figure 1.  Mechanistic investigations.  a  NMR kinetic evolution of the conversion of   3a  into  1a  and  2a  in function of time(solvent: deurated 1,1,2,2-tetrachloroethane (d 4 -TCE), zoom in 5–10 ppm, in Y axis: min incremented at the beginning of theheating at 336.9 K).  b  Eyring plot in d 4 -TCE from kinetic studies for the transformation of   3a  into  2a  performed at 324.1,330.2, 331.0, 333.6 and 336.9 K.  c  Isotopic labelling experiments and effect of the relative configuration (allyl chain/methylenealcohol) on the reaction outcome.  d  Kinetic evaluation of the production of aldehyde  2a  (conditions: 1 mol/L substrate in 3-pentanone measured by GC (conditions: 1 mol/L substrate in 3-pentanone in the presence of 5 mol% of   4 , 75 8 C. The reactionwas performed on a 0.8 mmol scale with decane as internal standard). red dots: conversion of   3a syn  into  2a  in function of time;blue squares: conversion of   3a d2  into  2a d2  in function of time; green triangles: conversion of substrate  3a anti  into  2a  in functionof time. COMMUNICATIONS  Adv. Synth. Catal.  2018 ,  360 , 1389–1396 1393  2018 Wiley-VCH Verlag GmbH &  Co. KGaA, Weinheim
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks